Private Investment Behaviour and Trade Policy Practice in ... · PRIVATE INVESTMENT BEHAVIOUR AND TRADE POLICY PRACTICE IN NIGERIA 3 of the most important factors determining the

Private Investment Behaviour andTrade Policy Practice in Nigeria

By

Dipo T. BusariUnited Nations African Institute for

Economic Development and Planning (UNIDEP)Dakar, Senegal

and

Phillip C. OmokeDepartment of Economics

Ebonyi State University, Nigeria

AERC Research Paper 177African Economic Research Consortium, Nairobi

April 2008

PRIVATE INVESTMENT BEHAVIOUR AND TRADE POLICY PRACTICE IN NIGERIA 1

1

1. Introduction

As part of the effort by many African countries to regain economic and financialbalance, government officials have increasingly turned their attention toimproving trade policies. These include actions on the exchange rate, import-

licensing, tariffs on imports, taxes or subsidies on exports, and institutional export-promotion measures (see Gunning, 1994, and Collier and Gunning, 1994, for detaileddiscussion on trade and trade policy issues in Africa under reform). How these changesaffect private investment is an interesting question for many African countries. Thedisappointing investment response in Nigeria, as in many sub-Saharan African (SSA)countries, to adjustment reform measures has brought to the fore in economic policydebate the role of private economic agents' perceptions about the credibility andsustainability of the reform measures.1 There is a large body of cross-country evidencethat more outward-oriented developing economies have tended to have higher investmentrates and per capita growth rates than less outward-oriented ones (Dollar, 1992; Edwards,1993; Ng and Yeats, 1997; Harrison, 1996; Chen and Funke 2003). Levine and Renelt(1992) find that the investment share of output is robustly correlated with variousmeasures of openness for a large sample of countries.

Trade liberalization is a core component of the structural adjustment programme(SAP).2 This entails the removal of all forms of quantitative restrictions on imports,reduction or elimination of tariffs, and so on. The correlation between openness andgrowth has tended to act as a major incentive for many countries to liberalize foreigntrade. Furthermore, since it is part of the larger SAP, trade liberalization is a preconditionfor enjoying many of the lending facilities of prominent multilateral lending institutionslike the World Bank and the International Monetary Fund (IMF). In the belief that moreopen economies perform better, international lending and donor agencies have attemptedto push developing countries in the direction of greater outward orientation by reducingrelative price distortions in the domestic market and promoting trade liberalization.However, the results of these programmes have been somewhat disappointing (seeFielding, 1997, and Collier and Gunning, 1999, for detailed discussions on the poorperformance of African countries). Detailed assessment reveals a pattern of at best weaklypositive effects on growth and a tendency to a negative impact on investment, althoughexport response has typically been strong for some countries (Corbo and Fisher, 1992;Greenaway and Morrissey, 1992; Mosley et al., 1991).

These findings, which appear to be at odds with the results of cross-countrycomparisons of growth and investment, form the main motivation for this study.Investment represents an area in which the impact of trade policy reform seems to be

2 RESEARCH PAPER 177

weakest in SSA (Chibber et al., 1992; Bleaney and Greenaway, 1993; Hadjimichael andGhura, 1995; Hadjimichael et al., 1995; Collier and Gunning, 1999).3 Many SSA countrieshave implemented far-reaching trade reform measures and are now re-introducing someprotectionist policies of different magnitude. In particular, many trade liberalizationattempts have been reversed because of the occurrence of severe and unexpectedmacroeconomic imbalances. The current democratic regime in Nigeria is implementingsome protectionist measures despite the fact that it is making serious attempts atliberalizing trade and attracting foreign investment.

The basic question is that if, by international trade theory proposition (and given theencouraging results of cross-country regression analyses), free trade is optimal, why aremany African countries re-introducing protectionist policies and what are the implicationsof such policy shift for capital accumulation? Recently, several authors have argued thatone possible explanation for this puzzle (i.e., the poor or even outright negative responseof private investment despite massive market-friendly reforms) is the lack of credibilityregarding the sustainability of trade liberalization. It is argued that trade policy reformscan be aborted and temporary protection programmes can become permanent becauseprivate economic agents do not perceive the current or future liberalization to besustainable, hence non-credible (Burnetti et al., 1997; Fielding, 1997; Collier andGunning, 1994; Collier and Patillo, 2000). In other words, the expectation that the reformwill be short-lived can in itself be self-fulfilling. The argument, for example as forwardedby Calvo (1987), is that the belief that a liberalization is just temporary may nourish animport rush – mainly of durable goods – because of the perceived low intertemporalrelative price of current import spending. Pressure will then grow in the country’s externalposition and, if there are limited international reserves (as in many SSA countries) orconstraints on external borrowing, a balance of payments crisis might develop. Hence,the government will re-impose trade barriers in order to preclude – or at least delay – thepolitical consequences of a painful devaluation. The point stressed here, and one thatshould be noted, is that pressure on external balance (and/or acute macroeconomicimbalance generally) could trigger a reversal in trade policy. Hence, the external balanceposition of the country can signal to private investors the sustainability of a currenttrade policy regime. This issue is central in this current study.

Thus, uncertainty regarding the sustainability of the trade liberalization mightdiscourage private investment and the allocation of resources. In particular, it is arguedthat if the sustainability of the liberalization is uncertain, the entrepreneurs will delayinvestment decisions because they do not want to commit resources to a particular sector(Bertola and Caballero, 1994; Rodrik, 1991; Dixit and Pindyck, 1994; Coy 1999).Consequently, capital accumulation might slow down, hindering economic growth, andthe allocation of resources away from the import-substituting sectors and toward theexport-oriented activities will not take place.

Stating the problem

The success of a structural adjustment programme (SAP) in bringing about asustainable recovery in economic activity in a given economy depends crucially on

the behaviour of investment in the aftermath of the reform process. In other words, one


of the most important factors determining the potential success of reform programmesis the extent and pace at which private investment responds to the policy changes. Sincethe expansion of public investment is usually constrained as part of fiscal austeritymeasures embodied in a SAP, the required recovery of investment has to come largelyfrom the private sector. The behaviour of private investment has therefore been a majorfocus of attention in assessing the reform outcome. The existing evidence across a widespectrum of developing countries generally points to a decline or stagnation of privateinvestment during the immediate post-reform years (World Bank, 1988; Harrigan andMosley, 1991; Greenaway and Morrissey, 1992; Gunning, 1994; Collier, 1995; Dehn,2000; Lemi and Sisay, 2001). The perceived risk of investment can be reduced in severalways. One is consistent, credible policy formulation that minimizes the likelihood ofpolicy reversal and maximizes its predictability. Credibility is crucial to capitalaccumulation, growth and overall successful macroeconomic policy. Confusing signalsmust be avoided, both to the public and among policy makers.

The empirical analysis of credibility issues relating trade and exchange rate policyreforms to private investment has been lacking for Nigeria. The expected investmentboom after the SAP commenced in 1986 seems not to have materialized. Privateinvestment share of the GDP is still below 10%. There was an initial rise in privateinvestment share of GDP just after the SAP was adopted, but the ratio has since declined.The country continues to implement far-reaching trade policy reforms with the hopethat private investment share will improve and non-oil exports will boom. The resultshave been disappointing. Studies have been carried out to examine the determinants ofprivate investment behaviour in Nigeria. Here too, the results have been controversialand hence inconclusive. Attention has been focused on the traditional determinants ofprivate investment such as output, relative prices, credit/liquidity and so on. It isinteresting to note that domestic credit to the private sector has continued to expand,and relative prices tend to favour investment in such sectors as agriculture andmanufacturing. However, the expected investment associated with such favourableenvironment has been elusive. It seems that some other factors are driving the responseof the private sector to investment spending beyond relative prices and currentprofitability.

It is then important to examine the perception of the private sector regarding thecredibility of the reform measures. This may well explain why, despite the far-reachingreforms implemented, private investment has responded unimpressively. For policypurposes, then, it is important to know how private perception of trade policy reformaffects investment. This will help in the design of more appropriate strategies to stimulateprivate investment.

Objective of the study

The study investigates if there is any empirical support for the credibility submissionin the literature in terms of fixed corporate investment in Nigeria. In this context,

the present study aims at examining the determinants of firm-level private corporateinvestment in Nigeria with emphasis on the trade cum exchange rate policy packageimplemented since 1986. The study does not construct a measure of trade policy


credibility directly. However, estimated coefficients on measures of macroeconomicinstability and uncertainty are used to interpret the notion of credibility in this study. Itshould be noted that there are several limitations attached to this indirect approach atmeasuring the impact of policy credibility on investment. The obvious one being thatmacroeconomic instability and uncertainty can be caused by other factors such as unrest– as it is being witnessed in the Niger Delta area of Nigeria – without necessarily leadingto credibility crisis. Such limitations are taken into account as we interpret our results.

Motivation for the study

Several studies exist on the determinants of private investment in Nigeria.4

Interestingly, many of these studies were undertaken in the last one decade or so.Many of these studies have generally used the traditional determinants of investmentfound in the literature and have focused on aggregate private investment. Whereas thetheory of investment is basically about the investment behaviour of the firm (or privatecorporate business investment), the dependent variable commonly used in the empiricalanalysis is total private investment, which includes both residential investment byhouseholds and investment by firms. Total private investment is a poor proxy for businessinvestment because household investment tends to dominate the total private investmentseries in most developing countries, and there are fundamental differences betweenhousehold residential investment (which is the key component of household investment)and business investment in terms of the underlying determinants.

Furthermore, none of these studies has attempted to examine the expectation/perception of the private sector in terms of the credibility of policy reform as a determinantof corporate private investment at the sectoral/firm level. This may provide the clue tothe puzzle about the disappointing response of corporate private investment to policyreform. Cross-country regressions tend to suggest that trade policy reform andliberalization are correlated with capital accumulation and hence growth (though thedirection of causation is still controversial). However, despite the massive reformmeasures introduced since 1986, the response of private investment has beendisappointing, thus contradicting results from cross-country regressions. What are thereasons for such contradictions? This question motivates this study. We seek to examinethe role of credibility in this contradiction. It should be observed that this study does notmeasure the direct impact of credibility, however, it interprets coefficients onmacroeconomic instability and uncertainty in this respect. Furthermore, methodologically,we deviate from the approaches adopted in previous studies on Nigeria.

The remaining part of this study is divided into five sections. Section 2 characterizestrade policy regimes in Nigeria and examines the investment response over the differentregimes. Section 3 presents a brief review of relevant aspects of the literature, whileSection 4 presents the methodological framework. The results from the analysis and thepolicy implications of such results are presented in Section 5 and Section 6 concludesthe study.


2. Characterizing Nigeria’s trade policyregimes

In this study we identified three broad trade episodes in Nigeria.5 The first correspondsto a period in which – except for the fixed exchange rate policy – trade policy interms of tariff measures and quantitative restrictions could be described as virtually

absent. In other words, the structure of tariff did not change over the time period. Duringthe second quantitative restrictions became a prominent tool with exchange rate policybeing absent. The final episode is the liberalization era that started with the adoption ofthe SAP. These are discussed in turn.

Period of highly regulated exchange rate policy: 1960–1974

At independence, Nigeria’s trade policy was virtually nonexistent, a trend thatcontinued until about 1967 before the outbreak of the civil war. However, exchange

rate measures represented a major policy instrument for controlling trade. This was interms of allocation rather than in terms of pricing. By 1970, due to the effects of the civilwar, there was cause for concern over the state of the country’s external payments position.Protection of the external payments position through prudent fiscal management, aswell as by enlargement of foreign exchange receipts, was therefore a major objective ofpolicy in the 1970/71 budget. Several measures were adopted. These included theimposition of additional indirect taxes on luxury commodities in high demand; relaxationof foreign exchange controls in respect of repatriation of dividends/profits andmanagement fees; and announcement by the government of its intention to renegotiatethe petroleum profits tax agreement. Among other measures were granting of taxexemptions to exporters of manufactured goods in respect of value or volume at a definedlevel, in order to encourage Nigerian manufacturers to export locally manufacturedgoods; and scrutiny of import valuation to discourage over invoicing of imports.

Following the suspension of dollar–gold convertability on 15 August 1971 Nigeriaadopted a new system of exchange rates with effect from 23 August 1971. Foreignexchange transactions were classified into two broad categories: contracts denominatedin US dollars, and contracts dominated in sterling. Also, the CBN maintained a fixedbuying and selling rate for the naira. Hence, until 1974 exchange rate policy was themajor instrument for controlling international transactions. The level of tariff was quitehigh, averaging about 47% in the early 1970s (see Figure 1). The level of economicopenness (the sum of export and import as a ratio of GDP) was also low (Figure 2).However, investment ratio (the ratio of private investment to GDP) was much higher(see Figure 3).

5


Figure 1: Average tariff rate

0%

30%

60%

70 74 78 82 86 90 94 98 02Years

Perc

enta

ge

Figure 2: Ratio of aggregate investment to total and non-oil GDP

0.0

0.1

0.1

0.2

0.2

70 73 76 79 82 85 88 91 94 97 00

Years

Rat

io

Inv_GDP Inv_nonoilGDP

Figure 3: (Export + import) as a ratio of GDP

0.00

0.30

0.60

0.90

1.20

70 74 78 82 86 90 94 98 02

Years

Rat

io


Era of quantitative restrictions and exchange controls:1975–1986

In 1975, in order to protect the two new motor car assembly plants in the country,government placed a quota on the importation of cars under 2000cc, representing the

difference between demand for such cars and their local production. This marked thebeginning of the use of quantitative and stricter tariff measures to control internationaltrade. Import duty reliefs were granted to the extent that was judged not to jeopardizethe competitiveness of local industries after taking account of other reliefs in the formof reduction in excise and company profit taxes. Some trade policy measures adoptedbetween 1976 and 1978 were designed to help decongest the seaports, conserve externalreserves and moderate imported inflation. The policy of appreciating the external valueof the naira was continued during these years. Controls were introduced on remittances.

The heavy drain on the nation’s external reserves in 1981, following excessivedisbursements of foreign exchange on importation at a time when foreign receipts weredeclining, called for a tightening of foreign exchange measures and imposition of tariffs,thereby raising average tariff (see Figure 1). Owing to the dangerously low level ofexternal reserves, certain medium and long-term and exchange control measures wereintroduced, aimed at a structural adjustment and diversification of the country’s sourcesof foreign exchange earnings. The trade policy component of the stabilization measurewas essentially exchange control measures. In 1983, under the Economic Stabilization(Emergency Provision) Act of April (as amended in November 1982), Nigeria undertookcomprehensive exchange control measures reducing the country’s foreign exchangeexpenditure to a level that would be compatible with its reduced foreign exchangecapacity.

Generally, these measures were pursued till 1985 with slight modifications by themilitary junta that seized power towards the end of 1983. During this period, the ratio ofprivate investment to GDP had nose-dived considerably. The level of economic opennesswas not particularly different from the previous period. Average tariff was much lower.

The era of economic reform and liberalization: 1987–2004

Because of the deteriorating state of Nigeria’s economy, particularly the externalreserve position (which could barely finance two months of imports) and the general

scarcity of commodities (then popularly called essenco – to mean “essentialcommodities”), the new military junta that seized power in 1985 decided to adopt theWorld Bank/IMF prescribed reform measures popularly called the structural adjustmentprogramme (SAP). This involved moves towards a more market-friendly trading systemand the dissolution of commodity marketing boards. Major trade policy reforms couldbe described as commencing in 1987. The two-tier foreign exchange market in operationfrom September 1986 was brought to an end on 2 July 1987 with the establishment ofthe unified foreign exchange market (FEM). The Dutch auction bidding system wasintroduced by the Central Bank of Nigeria (CBN). Certain trade policy measures werealso taken to strengthen the external sector.


The major aim of trade policy between 1987 and 1990 was, as in the previous years,to reduce pressures on the external sector. Hence, all the liberalization/deregulationmeasures adopted in 1988 to accomplish this objective were also retained in fiscal 1989.Early in 1989, both the CBN-based auction and autonomous market rates were mergedinto a single inter-bank foreign exchange market with a single naira exchange rate. Themerger was designed to stabilize the naira exchange rate and bring about more rationalbehaviour in the operations of the market and generally enhance its efficiency. Banksare expected to sell foreign exchange to their customers at not more than 1% above theofficial rate. In order to enlarge the scope of the official market for foreign exchangetransactions, the government approved the establishment of bureaux de change in 1986to be operated by private entrepreneurs. The daily bidding for foreign exchange by theauthorized dealers continued until 13 December 1990. On 14 December 1990, the Dutchauction system of bidding for foreign exchange was reintroduced in order to inject someelement of competition into the market. Emphasis continued to be on orderedliberalization of the foreign exchange market as well as active implementation of allexisting measures designed to diversity source of foreign exchange earnings other thanthe oil sector.

The thrust of trade policy between 1987 and 1999 was orderly liberalization andguided deregulation of the foreign exchange market under a close monitoringarrangement, as well as strengthening of the external reserve position so as to effectivelydefend the value of the naira against pressure from both the domestic and external fronts.The attainment of external balance was the focus of trade policy between 1992 and1999. Policy measures were directed at strengthening trade liberalization, the furtherderegulation of the foreign exchange market, and the pursuit of effective external debtmanagement strategies. As a result of the pressures in the foreign exchange market, theofficial segment of the market was further deregulated on 8 March 1992, by therealignment of the official exchange rate of the naira with the rate in the parallel market.At the same time, the practice of allocating foreign exchange to banks on predeterminedquotas was discontinued. Furthermore, the CBN was to enter the foreign exchange marketas an active participant, buying and selling foreign exchange from licensed foreignexchange dealers at the going market rate, with a view to influencing the exchange rateof the naira. The restriction on capital transfer was also abolished. Since 1999, the goalsof trade policy have been to maintain external balance and attract investment into thecountry. Various tariff measures have been put into practice. In recent times quotas andbans have been re-introduced, representing a policy shift from the practice of aggressiveliberalization.

Since 1995, access to foreign exchange at close to market rates and the lifting ofmost restrictions on current and capital transfers have significantly improved the tradeand investment environment. Nigeria became a founding Member of the World TradeOrganization (WTO) on 1 January 1995, following ratification of the WTO Agreementon 6 December 1994 by the Head of State. Nigeria is a signatory to the Lomé Conventionbetween the European Union and developing countries of Africa, the Caribbean and thePacific area (ACP). According to the Convention, Nigeria is granted duty-free access tothe EU market for exports of all industrial products and agricultural products that arenot subject to a common market organization in the framework of the EU’s Common


Agricultural Policy. Nigeria is the largest ACP exporter to the EU. Nigeria also benefitsfrom the Global System of Trade Preferences (GSTP). Major import liberalizationundertaken in 1995 significantly reduced tariff rates and reliance on quantitativerestrictions.

A pre-set tariff schedule, introduced in 1995 and valid until 2001, was intended tofurther decrease existing tariffs and reduce uncertainty for firms. Import liberalizationhas been pursued to reduce significantly the reliance on quantitative restrictions. Onlyad valorem tariffs were used in the new pre-set schedule. Import duties consisted of abasic rate of customs duty modified by an annually set rebate, plus a 7% surcharge. The1995–2001 tariff structure was designed to stimulate competition and efficiency byreducing tariffs on consumer items relative to tariffs on raw materials and intermediateand capital goods. The reduction of tariffs on final consumer goods was expected toexpose domestic manufacturers to import competition, while the relatively higher tariffson raw materials were supposed to attract investment into raw material and intermediategoods production. In the course of the reform programme, all excise duties levied ondomestically produced goods were removed in January 1998. By 2000, most bans onimports were abolished and by 2004 many goods that were hitherto unbanned cameunder ban again.

In this episode, we observed that average tariff had dropped significantly and thelevel of economic openness had reached an all time high of over 90%. Although privateinvestment seems to be on the upward direction, it is still below the ratio in the 1970sand highly variable (see figures 1 to 3).


3. Review of the literature

Investment literature abounds with descriptions of the determinants of privateinvestment and the channels through which such variables affect investment.Prominent amongst these are the traditional neoclassical theory, as formulated by

Jorgenson (1963, 1971), which postulates the role of the cost of capital; the acceleratormodel, which postulated the role of rate of change of output; the Tobin's q, which arguesa role for the value of the firm; and the financial repression framework due to McKinnon(1973) and Shaw (1973). Over time more variables have been observed to affect corporateinvestment in one way or the other. This study will not attempt to repeat the literature inthis respect. However, interested readers are referred to studies like Greene and Villanueva(1991); DeLong and Summers (1991); Chhibber et al. (1992); Serven and Solimano(1993); Bleaney and Greenaway (1993); Bleaney et al. (1995); Ibarra (1995); Bleaney(1996); and Blomstrom et al. (1996). Levine and Renelt (1992) made a classic discussionof the relative impotence and sensitivity of many of such variables. However, during the1990s, another strand of argument on the determinants of private investment started togain ground as coherently formulated by Pindyck (1991). This new line of argument inthe recent literature on investment interprets a firm as consisting of a portfolio of options,and uses options-based pricing techniques to study the investment decision.6 We believethat this line of argument is more relevant in the current study and hence attention isfocused on it in the following section.7

The real options theory of investment

Using options-based pricing techniques to study the investment decision of firms,the real option theory of investment interprets a firm as consisting of a portfolio of

options. As argued by Chen and Funke (2003), investment opportunities can be viewedas “option-rights” such that each investment project can be assimilated, in its nature,into the purchase of a financial call option, where the investor pays a premium price inorder to get the right to buy an asset for some time at a predetermined price (exerciseprice), and eventually different from the spot market price of the asset (strike price).

In a similar manner, while making investment decisions, a firm pays a price (the costof setting up the project) giving it the right to use the capital (exercise price), now or inthe future, in return for an asset worth a strike price. The basic implication of this analysisis that the wholesale application of the net present value rule to the expected future cashflows of the firm will give suboptimal results (Chen and Funke, 2003). To avoid this

10


suboptimal investment decision rule, it is important to consider the following threecharacteristics of the firm’s investment decision:• There is uncertainty about future payoffs from the investment;• The investment does not entail a now-or-never decision; and• The investment is at least partially irreversible.

As argued in the literature, the direct implication of the foregoing characteristics offixed corporate investment for optimal investment decision making is that the opportunitycost of investment will necessarily include the value of the option to wait that isextinguished when an investment decision is taken (Abel and Eberly, 1994; Abel et al.,1996). Hence, Chen and Funke (2003) argue that the investment decision is affected bythe determinants of the value of the option; consequently, an appropriate identificationof the optimal exercise strategies for real options plays a crucial role in the maximizationof a firm’s value. The real options studied in the literature include, among others, operatingoptions (McDonald and Siegel, 1985), the option to wait and undertake an investmentlater (McDonald and Siegel, 1986), and uncertainty from future interest rates (Ingersolland Ross, 1992). Other contributions to the literature are Abel and Eberly (1994, 1997)and Abel et al. (1996).8 The general focus in the literature has been the effect of demand,price and/or exchange rate uncertainty upon investment decisions of firms. On the basisof the objective and focus of this study, we now review the relevant aspects of the realoption theory to the macro-policy environment/uncertainty.

Trade policy reform, investment and economic performance

Many reasons why trade liberalization might encourage investment are illuminatedin the literature. Corden (1974) pointed out that protection could reduce the rate

of capital accumulation because, in the absence of capital flows, investment is determinedby the amount of domestic savings out of total income. Hence, as long as protectionlowers real income, especially for small countries that lack international market power,investment and the rate of capital accumulation will decline. The new literature on economicgrowth argues that countries that take advantage of international trade might enjoy highergrowth because of faster absorption of foreign technical knowledge (Grossman andHelpman, 1991; Aizenman, 1992; Aizenman and Marion, 1993; Romer, 1994).

In other words, if investment is linked to changes in output (say through the acceleratoreffect), any policy measure that promotes growth will be a stimulus for an increase incapital accumulation. In an indirect manner, it is argued that in as much as economicgrowth is linked to faster capital accumulation and growth is associated with openness,then effective trade policy reform will be investment inducing (see OECD, 2001; Rodrik,1997; Winters, 2001). There is a preponderance of cross-country evidence that tradeliberalization and openness to trade increases capital accumulation, the growth rate ofincome and output (see Sachs and Warner, 1995; Dollar, 1992; Edwards, 1993, 1998;Ben-David, 1993; Frankel and Romer, 1999). In addition, numerous individual countrystudies over the past three decades suggest that “trade does seem to create, even sustainhigher growth” (Bhagwati and Srinivasan, 1999: 8).


It is generally agreed, however, that a country’s trade policy is the key link in thetransmission of price signals from the world market to the national economy. The literaturesuggests that the existence of uncertainty regarding the sustainability of tradeliberalization might discourage investment and deter resource (re-)allocation. This isusually the case when investors are risk-averse (see Arellano, 1990; Dixit and Pindyck,1994; Hassett and Metcalf, 1999; Collier and Patillo, 2000). More recently, it has beenshown that risk-aversion is not the only way to generate this conclusion: an alternativeis provided by the literature on irreversible investment and hysteresis, which shows thatthe existence of sunk costs, in combination with price uncertainty, might frustrate theuse of relative prices as a guide for investment decisions. This lack of credibility regardingthe durability of policy reforms on the one hand, and insofar as physical capital is partlyor fully irreversible on the other, even if investors are risk-neutral, they tend to favour a“wait-and-see” attitude in the early stages of a reform programme (Rodrik, 1989a/b;Bassanini, 2006; and Rajan and Marwah, 1998).

Given this irreversibility of physical capital, it may pay the investor to delayinvestments until uncertainty is reduced, even though the postponement option carries apremium. This premium, which gives the investor the right but not obligation to enterthe market some time in the future – akin to a financial call option – is either implicit interms of loss of market share and/or forsaking other advantages of being a market pioneer,or explicit in terms of initial sunk costs to preserve the advantages it may have fromearly entrance. Thus, as Metcalf and Rosenthal (1995: 521) state, “part of the cost ofmaking an investment is the value of the option that is lost when the option is killed”.

That is, from the firms’ point of view, it might be optimal to delay expansion ofproductive capacity even if the actual price exceeds the long-run average cost–that isthere is an option value in holding back investment decisions (Dixit and Pindyck, 1994).The reason is that when investment decisions are irreversible and relative prices areuncertain, waiting represents an opportunity cost that should be compared with thecurrently available profit in order to determine the optimal timing of investment (seePindyck, 1991, and Chen and Funke, 2003, for a survey of this and related issues). Fordeveloping countries, Rodrik (1991) used this framework to show that private investmentcan fail to take place after the introduction of an economic policy reform if the survivalof the policy is uncertain or perceived to be unsustainable by private economic agents.

In essence, there are two fundamental reasons for the observed link between tradeand overall economic performance. The first is that both depend on the policies andinstitutions in place in the transition country and its major trading partners. The secondis that increased openness to the world appears to have a strong impact on rates ofeconomic growth (Frankel and Romer, 1999). The first source of the observed correlationhighlights the importance of reforming trade policy as part of a more comprehensivepackage of reforms aimed at achieving economic development. The second highlightsthe importance of reforming trade policies if the reforms are to succeed in raising livingstandards and alleviating poverty.

While firms obviously face many more complicated conditions and trade-offs,expectations and perceptions of policy uncertainty do lead them to exercise some realoptions in their decision making processes. Taking the discussion above to its logicalend leads one to conclude that at the extreme, uncertainty about future policy reversals,by “crippling animal spirits” of the private investors (Rodrik, 1989a: 3), may – throughthe multiplier-accelerator mechanism – leave the economy trapped in a “low investment


equilibrium”. Recognizing this, Rodrik (1990: 934) has concluded that from the viewpointof investment, “liberalization may often need to take a back seat when it places thesustainability of policies...into question”.

Macroeconomic policy reforms: Credibility, reversal,and investment behaviour

From the policy viewpoint, an extremely important form of uncertainty faced byinvestors is the imperfect credibility of policy reforms. Investment-friendly reforms

typically raise expected returns, but may also increase uncertainty if investors believethat the reform measures could be reversed. In such a context, investors’ perceptionsabout the probability of policy reversal become a key determinant of the investmentresponse. These issues are explored by Rodrik (1991) using a model in which investmentinvolves sunk costs of entry and exit. He shows that a reform favourable to capital, butregarded as less than fully credible, will fail to trigger an investment response unless thereturn on capital becomes high enough to compensate investors for the losses they wouldincur should the reversal take place. Similar qualitative conclusions are reached by vanWijnbergen (1985) who considers the case of a trade reform suspected to be onlytemporary. An economic (reform) policy enjoys credibility to the extent that relevantactors, such as domestic and foreign investors, believe the government will implementand sustain the programme of reforms that it has announced. The identity of relevantactors may vary across time and space, but the issue of credibility seems inescapable,given the sequential nature of economic decision making. At least in principle, agovernment that dismantles protectionism today can restore it tomorrow, just as agovernment that cuts taxes now can escalate them later. The record of trade liberalizationin developing countries is replete with examples of governments that promised onepolicy but delivered another, or implemented reforms that were subsequently retracted(see Michaely et al., 1991, for a detailed discussion of this issue).

If investors doubt the longevity of free trade, for example, they may decide not toshift resources from inefficient, import-competing industries to more dynamic, export-oriented ones. The deterrence to investment arises because exporting involves costs thatwould be difficult to recover if the government reinstated protectionism. For example,physical capital is typically expensive to install and uniquely appropriate to a particularindustry. Likewise, investments in human capital (hiring and training) perform best inthe activity for which they were designed. Firms will avoid making export relatedinvestments in client networks and physical and human capital, unless they believe thatpublic authorities will persist in keeping the economy open. If investors anticipate apolicy reversal, then commercial liberalization will hurt import-competitors withoutstimulating the growth of exports, and the liberal policy will become unsustainable(Rodrik, 1992).

Sources of policy credibility crisis

Several sources of credibility problems are identified in the literature (for a recentdiscussion of this issue see Aizeman, 1992; Drazen and Masson, 1994; Agenor and

Masson, 1999; Aizenman and Marion, 1999). One is due to the problem of time-


inconsistency. This is a situation in which the government has incentives to deviate ex-post from the optimal trade policy announced ex-ante. In this case private economicagents take into consideration such incentives (see Eaton and Grossman, 1985, andStaiger and Tabellini, 1987, for models in which tariffs are used to redistribute incomewhen there are uncertain terms of trade shocks and some factors are fully or partiallyimmobile). Another source of credibility is the level of political uncertainty. That is,situations in which private economic agents do not know whether trade liberalizationwill still be an objective of the government in the future. In some cases, the probabilityof reversal depends on the information held by private economic agents concerning thetrue intention of the government and in other cases, this probability depends on thelikelihood of a political party (or regime) that does not support the liberalization to beelected (Froot, 1988, and Engel and Kletzer, 1987, have adopted models in which, afterthe start of a trade liberalization, the private agents update continuously the probabilityof reversal). Hence, a current liberalization effort is perceived not to be entirely crediblesince there is always a chance that it will be reversed by a future protectionist government.Another source of trade policy reversal is the collapse of the balance of payment.Expansionary fiscal and monetary regimes, declining trends of the real exchange rate,external shocks to the economy, and weak export performance are among the factorsthat usually lead to a deterioration of a country’s external position and, frequently, to theabortion of trade reforms.

The last source mentioned needs further investigation. The lack of sustainability oftrade liberalization might arise from the inconsistency with other accompanying economicpolicies. For example, an expansionary fiscal policy and a fixed exchange regime arelikely to induce deterioration in the balance of payments. Hence, to prevent the depletionof international reserves or to avoid politically painful devaluation, the authorities areusually inclined towards re-imposing trade restrictions. In a like manner, the sustainabilityof trade reform might be in doubt when the government attempts, simultaneously, thestabilization of the economy and the liberalization of trade.

The stabilization effort might require the fixing of nominal exchange rate in order touse it as an anchor for the nominal variables. However, it is observed that if such policyleads to an over-valuation of the real exchange rate, the odds for a successful liberalizationare then largely diminished. Hence, in the absence of capital flows, the deterioration inthe current accounts might lead to a reversal of the reform effort (Ibarra, 1995). Also,trade liberalization might be reversed as a result of negative external shocks that cannotbe adequately handled by the government facing binding restrictions on internationalborrowing. This is usually the case if the country depends on one or a few major productsfor most of its export earnings (as Nigeria depends largely on oil earnings). Thegovernment will try to cut imports – raising tariffs and re-imposing quotas – to preventthe collapse of the exchange rate or the drainage of foreign exchange reserves. In sum,trade liberalization is less credible when the government pursues inconsistentmacroeconomic policies or, in general, when there is a high probability of occurrence ofa balance of payment crisis (this issue is central in this study).

Of course, trade policies can fail by themselves, if they are poorly designed, but theempirical evidence available to date indicates that the absence of companion policies isthe principal reason for a reversal of trade liberalization. Companion policies includemacroeconomic policies as well as policies that affect the allocation of resources among


different sectors and markets. Exchange rate policies assume a crucial role in a tradeliberalization. This is because trade liberalization could lead to a trade deficit sinceimports respond faster than exports to a change in relative prices. Moreover, the situationprior to trade liberalization is usually one of an over-valued exchange rate maintainedwith import restrictions, thus a liberalization that allows the allocation of resourcesbased on the prices implies a change in fundamental macroeconomic price such as thereal exchange rate.9 Another reason from the trade practice point of view for policyreversal is the inability of developing countries to secure their export markets. This isusually the case when advanced countries adopt legislation that makes it difficult fordeveloping countries to export to such advanced economies. Hence, the export collapsecould initiate a reversal of trade policy. In sum, as McCulloch and McPherson (2002: 18)argued, the circumstances associated with policy reversals in SSA include (but are notlimited to):

external shocks, especially commodity price collapses, over-expansion during booms,slower than expected debt relief, resistance by central banks to the removal of exchangecontrols, delays in unwinding state-run trading monopolies, need for revenue holding uptariff reform, imposition of new non-tariff restrictions such as quality standards for imports,deterioration of physical infrastructure, especially related to transport, exchange rateappreciation due to donor aid flows, activism by ministries of commerce to “promoteindustrialization”, unwillingness of government officials to eliminate special entitlementsfor the military, inability to control or reduce smuggling, inability to reduce unit labourcosts, thereby improving competitiveness, rising protectionism in developed countries(real or potential), difficulties in penetrating regional markets, and incoherence of donorconditions, so that government operates at cross purposes.

Hence, reversals arise for a wide variety of reasons that may be technical, bureaucratic,institutional and often personal (to the leaders). These causes of reversals can convenientlyfit into any of the four sequential elements in the liberalization process: (1) defects ofpolicy design, (2) lack of political support, (3) defects of implementation, and (4)environmental problems that undermine policy effectiveness. This explanation issummarized as a schema in Figure 4.

Figure 4: A schema on credibility concept

Infrequent implementationPolicy Withincommitments Partial implementation reasonable time

scalePolicy reversal

Broad credibility crisis sources

• Dynamic (time) inconsistency: Incentives to deviate ex-post from the optimal trade policyannounced ex-ante (this includes design and implementation problems).

• Macroeconomic imbalances: For example, collapse of the balance of payment leading totrade policy reversal.

• Political interference (political uncertainty): Would trade liberalization still be an objective ofthe government in the future?


4. Methodological framework

Our modelling approach is to start first with a theoretical model, which we callthe core model. The core model follows the neoclassical line of thought thatinvestment spending is determined by two broad elements: the accelerator and

the user cost of capital effects (see, for example, Jorgenson, 1971; Ibarra, 1995; andAthukorala and Sen, 1996). The accelerator effect captures the relationship betweencapital accumulation and the rate of change of output. The user cost captures the degreeof substitutability between capital and other inputs.

The core (neoclassical) investment model of the firm

Using a two-factor model of investment behaviour in the tradition of the neoclassicaltheory – capital and labour – the firm maximizes profit subject to a technology that

can be represented by a CES production function – Y = γ(Kρ+ Lρ)v/ρ, where γ is theefficiency parameter, ρ is the substitution parameter and v is the returns to scale parameter.The desired stock of capital that results from the first order conditions can be expressedas follows:

K* = A[Y]φ[ωk /P]-σ (1)

where K* is the desired capital stock, Y is real output, ωk is the user cost of capitalservices, P is the output price, σ is the elasticity of substitution between capital andlabour, φ is the elasticity of the optimal capital with respect to output, and A is a scalefactor. The user cost of capital can be expressed as

ωk = Pk [(r-π)/(1+π)+δ-κ-τz]/(1-τ) (2)

where Pk is the purchase price of a unit of new capital; r is the nominal financial cost ofcapital, usually a weighted average of the external and internal cost of funds; π is therate of inflation; δ is the rate of depreciation of the capital stock; τ is the rate of corporateincome taxation; κ is the rate of investment tax credits; and z is the present value of taxdepreciation allowances. Taking the differential of the logarithmic transformation ofEquation 1, and introducing costs of adjustment using a distributed lag function of theKoyck type, it is possible to get an expression that relates current investment to pastinvestment, to rate of change of output and to the rate of change in the relative price ofcapital services. To transform these mathematical relations into an economic model of

16


investment behaviour, the actual rates of variation of output and the relative price ofcapital services are replaced by their expected counterparts. The result is the followingequation (see Ibarra, 1995 and Athukorala and Sen, 1996):10

It /Kt-1 = κo + κ1[It-1/Kt-2] + κ2Et[∆ln(Yt )] +κ3Et[∆ln(ωkt /pt)] + εt (3)

κ1, κ2 > 0; κ3 < 0.

Et[.] denotes the expectation operator applied to the information set available at thebeginning of time t, which is usually proxied by an autoregressive process.

Alternative model formulation: The Tobin’s q

It is generally argued in the literature that uncertainty affects both the timing and theamount of investment. In particular, uncertainty could cause firms to delay capital

expenditures – options to wait. The effects of trade policy may also be examined usinga model based on forward-looking optimization by firms. The Tobin’s q model of firminvestment is such a model.11 This model of investment was suggested by James Tobinin 1969 and formalized by Hayashi (1982). The q theory of investment states that q is asufficient statistic for summarizing all the information relevant to a firm’s investmentdecision. For empirical purposes, this q variable is usually defined as the ratio of themarket value of the firm to the replacement cost of its capital stock.

Empirical research typically rearranges the theoretical first order condition as a linearregression under the assumption of quadratic adjustment costs, and then uses theobservable Tobin’s q as a proxy for marginal q. Hayashi (1982) shows that under theassumptions of constant returns to scale and perfect competition, marginal q is equal toaverage q. If this ratio is less than or equal to unity, then there is no incentive for the firmto invest in plant and equipment. In this case, the firm’s shareholders could earn a higherreturn elsewhere. If the value of Tobin’s q is greater than unity, then the firm shouldinvest in capital goods in order to maximize the return to its shareholders. From amacroeconomic point of view, Tobin’s q is presented as the ratio of firms’ stock marketcapitalization to the replacement cost of their physical capital as follows:

K

Vqρ

= (4)

where V is the stock market capitalization of private firms and ρK is the replacement costof their physical capital. Hence, the core Tobin’s q investment model is specified as:

itit

itiit K

IqKI ξββα +⎟

⎠⎞

⎜⎝⎛++=⎟

⎠⎞

⎜⎝⎛

−1210 (5)

where I/K is the rate of capital accumulation and the lag refers to the investment processlags, q is Tobin’s q, and αoi the intercept(s). The residual ξ is usually assumed to be


identically and independently distributed. It is common to add a measure of cash flow,defined by the ratio of earnings before interest and taxes (EBIT) to capital stock, toEquation 5.

The extended investment model

For the purpose of this study we augment these basic (core) models by adding sets ofother explanatory variables (see Table 1 for the description of the variables). It

should be observed that numerous factors affect investment in theory. Levine and Renelt(1992) argue that these factors are well over fifty. However, a common feature of moststudies on investment regressions is that the explanatory variables are enteredindependently and linearly. This approach is followed in this study. The choice of mostof these variables is quite standard in recent literature on determinants of investment.The role of the traditional determinants may be found in the review by Everhart andSumlinksi (2001). On financial constraints variables, readers are referred to Gertler(1988), Chirinko and Schaller (1995), and Loungani and Rush (1995); and to Ferderer(1993), Pindyck and Solimano (1993), Serven and Solimano (1993), Price (1995), andServen (1996) for uncertainty variables. Bardhan (1984, Chapter 4) and Blejer and Khan(1984) are sources for the role of public investment, and Asiedu (2002, 2004) and Serven(2002) have studied trade openness and capital control variables. Some variables inTable 1 will require further explanation in the context of this study; this is done insubsequent sections.

Table 1: Description and measurement of the explanatory variablesCF, • Change in log of real output averaged for t-5 to t-1Conventional factors (accelerator effect)

• Change in log of cost of capital average for the period t-5 tot-1 (user cost of capital effect)

• Private sector credit as percentage of GDP average for t-5 tot-1 (proxy for financing constraint)

UF,Uncertainty/macro- • Standard deviation/coefficient of variation:economic (instability), • Real exchange ratepolicy environment • Inflation/interest rate/nominal money growthfactors • BOP

• External reserves• Fiscal variable (budget deficit)• Terms of trade• Output variability• (Proxy for macroeconomic uncertainty, uncertainty about

profitability of investment project and policy credibility forperiod t-5 to t-1)

PF, • Government expenditure on infrastructure as percentage ofPolicy related factors GDP average for t-5 to t-1 (complementary public investment

in sector i)• Capital controls (measures the restrictiveness on capital

market transactions–another proxy for openness)a

Continued


Table 1, ContinuedCR, • Real exchange rate depreciation average for t-5 to t-1Trade, trade policy, (relative price effect)exchange rate and • Ratio of export plus import to GDP averaged for t-5 to t-1credibility factors (measure of openness)

• Average tariff rate and non tariff measure on intermediateand capital goods, average for t-5 to t-1 (proxy for antiimport liberalization attitude of the government and restrictionson trade and investment)

• Standard deviation/coefficient of variation of tariff rate/non-tariffbarrier index average for t-5 to t-1 (proxy for lack of credibility ofimport liberalization efforts of the economy)

• Cost of import index (another proxy for openness of trade)b

• Black market premium, average for t-5 to t-1 (proxy for exchangerate control)

• Number of bilateral investment treaty between Nigeria and allother countries (proxy for policy\political commitment)

Dm, • Dummy variable for periods of membership in MultilateralCommitment dummy Investment Guarantee Agency (MIGA); takes value of 1 since

the year Nigeria signed the agreement (12 April 1988) and 0otherwise (proxy for policy commitment).

Notes:a. This measures restrictions on the freedom of citizens to engage in capital market exchange with foreigners- index of capital controls among 13 IMF categories. It ranges from 0 to 10, a higher rating implying fewerrestrictions (published by the Fraser Institute and available at www.freetheworld.com/). This corresponds toitem 4E:ii of the Economic Freedom of the World (EFW) index.b. Costs of importing: The combined effect of import tariffs, licence fees, bank fees and the time required foradministrative red-tape raises costs of importing equipment by (10 = 10% or less; 0 = more than 50%)(GCR). This is item 4B:ii in the EFW index.

Data and estimation

The data used in the analysis are described in detail in the Appendix. In this study,we will be estimating a regression equation of the form:

Invit=αι0+α1 Invit-1+α2CFit+α3UFit+α4PFit+α5CRit+α6Dmt+εit (6)

i=firm, t=1980-1984, 1985-1989, 1990-1994, 1995-1999, 2000-2003

where et = the error term and αi0 is an unobserved time invariant sector/firm specificeffect. The data are averaged for five years. Evidence tends to suggest that five-yearaveraging in panel data out performs estimates without averaging and it also reduces thepossibility of serial correlation in the error term (Ruggiero, 2003; Cruces et al., 2003;Pessoa et al., 2004).


Measuring volatility

The variables used in the measure of volatility are intended not only to capture tradeand trade policy related issues, but the entire macroeconomic policy environment.12

They are defined as follows:• Aggregate demand variable

Percentage real GDP growth• Fiscal variable

Fiscal deficit as a percentage share of GDP• Monetary variables

Percentage nominal money growth (M1)Percentage CPI inflation ratePercentage nominal lending rate

• External sector variablesPercentage change in the real exchange ratePercentage change in terms of tradeExternal reserve as a percentage share of GDPCurrent account balance as a percentage share of GDP

Different approaches, as contained in empirical literature, were adopted in the measureof macroeconomic volatility. These are:• Standard deviation of the variable over the specific period.• Conditional-heteroscedastic GARCH(1,1).• Forecast-error (actual minus linear trend predictions) standard deviation of residual.• Simple AR(1) order, standard deviation of residual.

Table 2 presents the correlation analysis between the various measures of volatilityidentified. It can be observed that measures a, c and d are correlated above 0.95. Onlythe GARCH(1,1) measure shows correlation below 0.75 with other measures. Henceany of the measures defined in a, c and d can be used. The correlations show that theyare close alternative measures. In this study, we adopt measure (a) for further analysis.This measure is also widely used in the literature in measuring volatility.

Table 2: Correlation of measures of volatilitya b c d

a 1.000b 0.665 1.000c 0.996 0.667 1.000d 0.954 0.715 0.956 1.000

Investment treaties and MIGA

Nigeria joined the World Bank Group Multilateral Investment Guarantee Agency(MIGA) in 1988. It is aimed, amongst other things, at multilateral risk mitigation

by promoting foreign direct investment into developing countries by insuring investors


against political or non-commercial risks. Coupled to this is the signing of bilateralinvestment treaties (BITs) with several countries. This initiative started in February1990 when an agreement was signed with France to take effect in 1991. By December1990 another agreement was signed with the United Kingdom and in November 1992an agreement with Netherlands was signed to take effect in 1994. Since then, Nigeriahas signed agreements with Germany, Belgium, South Africa, Italy, Argentina, Egypt,South Korea, China, Jamaica, Sweden, Switzerland, Turkey, Uganda and Romania. Whilea Trade and Investment Framework Agreement (TIFA) has been signed with the UnitedStates, a bilateral investment treaty is not in place.

Although the main focus of MIGA and bilateral investment treaties is to attract foreigninvestment, the initiatives are capable of sending strong signals to domestic investorsabout the seriousness and commitment of the government in establishing and sustainingan investment-friendly environment. In this study, we use the number of investmenttreaties as a variable and the year since membership of MIGA as a dummy variable. Therole of BITs in attracting investment is still contentious in the empirical literature, butwe are not delving into that in this study.

Import intensity and trade policy

To assume that trade policy reform would affect all firms equally could be toorestrictive an assumption. Firms have different export/import intensities, hence,

they are likely to react to (or be affected by) trade policies differently. Export intensitiesin Nigeria’s manufacturing subsector are generally low, below 10% on the average.Firms are likely to be more sensitive to policies that affect intermediate imports ratherthen exports due to high import intensity of firms. In this study we define import intensityas the share of value of imports of a firm in value of output at factor cost. There are atleast two approaches to incorporating import intensity into the analysis.

First, we could define a threshold above which a firm is regarded as having highimport intensity. Hence, a binary variable can be defined such that firms with low importintensity take a value of zero and those with high import intensity take a value of unity.In the literature it is common to assume that a value above 30% implies high importintensity. The second approach is to use the actual values of import intensities of thevarious firms. The latter approach requires having detailed information about inputrequirements of each firm. This is quite difficult, but a way out of this is to assign theindustry average to each firm in a particular industry. The import intensity values canthen be interacted with trade policy variables to examine if there is any differentialimpact on how trade policy affects firms. In this study, we adopt the first approach: theuse of binary variable to classify firms as import intensive.

Trade policy and inter-sector allocation of investment

First it is important to point out that one of the effects of trade policy, particularlytrade liberalization, on private investment would largely come about through a shift

of resource allocation in response to the changes in relative prices.13 Hence, an appropriate


way of analysing the behavioural response of private economic agents is to detect inter-sector changes in private investment. There are several ways of estimating inter-sectorchanges in capital allocation. With availability of a reliable input-output (IO) table thiscan be derived easily. This is not the case for Nigeria, however. Hence we follow theapproach of Galindo et al. (2002) to investigate whether trade policy (defined essentiallyas trade liberalization) has increased the share of investment going to firms with a highermarginal return to capital.

To achieve this, we calculate a summary index of the efficiency of allocation ofinvestment. It should be observed that in assessing the effect of trade liberalization wewant to see whether it succeeds in directing resources towards those uses with highermarginal returns. This is the narrow concept of efficiency we focus on. In measuringefficiency in the allocation of investment amongst sectors it is important to have a measureof the marginal return to investment. Galindo et al. (2002) suggest two approaches. Thefirst is based on the ratio of sales to capital, while the second is based on ratio of operatingprofits to capital. Based on data availability we chose the latter, that is, the ratio of salesto capital.14 To measure the efficiency of the allocation of investment in a year, each ofthe estimates of the total return on investment must be compared with a benchmark.Galindo et al. (2002) suggest that a reliable benchmark is an estimate of total return ifinvestment funds had been allocated to firms in proportion to their share of capital in theeconomy.

According to Galindo et al. (2002), in every case it is sufficient to use the sameestimates of the marginal product of capital for each firm to estimate actual returns, andreturns for the benchmark allocation. We divide our measure of total return actuallyachieved by this benchmark to obtain a measure of the efficiency with which investmentfunds were allocated in each year. Hence, efficiency of the allocation of investmentfunds (EI), where sales per unit of capital is used as a measure of the marginal productof investment, is given as:

∑

∑

+

+

+

+

=

i

TtT

t

ti

ti

ti

iti

ti

ti

st

IKK

KS

IKS

EI,

1,

1,

,1,

1,

(5)

where Sit denotes firm i's sales at time t. Iit and Kit are, respectively, firm i’s investment

and capital at the beginning of time t. T

tI and T

tK represent, instead, aggregate investment

and aggregate capital at time t, respectively. The expression assumes that investmentbecomes productive with a one-period delay. Furthermore, it uses an individual firm’scapital stock at the beginning of year t as a fraction of total capital for all firms at thebeginning of the same year to measure the proportion of investment funds that the firmwould receive if investment funds were assigned in the same proportion as in the past. Itshould also be noted that each unit of investment in year t is assumed to increase thecapital stock, and hence generates a return in year t+1.


Trade policy and sectoral classification

It should be expected that trade policy will affect firm-level investment differentlydepending on the type of activity, especially on the degree of exposure to international

trade (import of inputs and export of output). Trade policy can have opposite effects inthe input and output markets. For example, a firm producing non-tradeable goods butusing imported inputs will be negatively affected by exchange rate depreciation, whichwill cause an increase in the cost of production (input) without any gains on the revenueside, thus reducing incentive to invest.

For a firm that uses domestic inputs and exports its output, a depreciation of theexchange rate will induce investment. To capture this kind of relationship between tradepolicy reforms (particularly exchange rate reform), we classify firms into tradeable andnon-tradeable sectors. A dummy is then created, which assumes the value of unity if thefirm is non-tradeable and zero otherwise. For the purposes of our analysis and to avoidambiguity, the broad inclusions for the non-tradeable goods sector is as follows: hotelsand restaurants; transport, storage and communications; financial intermediation; realestate, renting and business activities; public administration and defence; compulsorysocial security; education; health and social work; electricity, gas and water supply; andother community, social and personal service activities. Others are classified as tradeablegoods.

Defining the real exchange rate (ε) as the ratio of the price of tradeable goods (ρT) tonon-tradeable goods (ρN) such that ε> 0 implies depreciation, then we can ask what doesdepreciation of ε do to profits (π) and cost of capital (ρk)? This will allow us to know thedirection of investment (I) in each sector when there is a policy change. Naturally, theprice of the tradeable sector is the international price of tradeable goods multiplied bythe exchange rate and adjusted for import tariff. All that is required to know the expectedbehaviour of investment in both the tradeable and non-tradeable sectors is to evaluatethe following (partial) derivatives (see Table 3.15):

TNTNKK

ρπ

ρπ

επ

επ

∂∂

∂∂

∂∂

∂∂ ;;;

Table 3: Expected impact of exchange rate depreciation on investmentGoods market

N T

Input π ↓ ρk ↑ π ↑ ρk ↑T I ↓ I ?

Market π ↓ ρk → π ↑ρk →N I ↓ I ↑Note: N and T imply, respectively, non-tradeable and tradeable goods sectors


From Table 3, it is postulated that when a firm belongs to the non-tradeable goodssector in terms of output but its (capital) input is tradeable, then a depreciation of theexchange rate would reduce profit and raise the cost of capital, thereby reducinginvestment. When the firms’ input and output are both tradeable goods, depreciationwould increase both profit and cost of capital. Investment direction is uncertain. Whenboth the output and capital input of the firm are non-tradeable goods, then we expectprofit to fall but cost of capital to be unaffected. Hence, investment will fall. Furthermore,when the output is a tradeable good and the input is a non-tradeable good, then weexpect profit to rise and cost of capital to be unchanged. Hence, investment will rise.The foregoing argument can be extended to deal with other aspects of trade policybeyond exchange rate adjustment, such as tariff adjustment. However, in this study, werestrict our analyses to exchange rate adjustment. The sectoral classification dummy isinteracted with the real exchange rate to evaluate the impact of exchange rate and sectoralclassification on firm investment in Nigeria.


5. Analyses and results

It has become conventional to view long-run parameters as reflecting cointegratingrelationships among a set of I(1) variables. The standard methodology in such casesis to first establish the order of integration of the variables in question, and then –

having established that the variables are of the same order of integration – test whetherthere is at least one linear relationship among these variables. There are several ways oftesting for unit roots in short panel data. These include Levin et al. (2002), Breitung(2000), Im et al. (2003), Fisher-type tests using augmented Dickey–Fuller (ADF) andPhillip–Perron (PP) tests (Maddala and Wu, 1999; Choi, 2001; Hadri, 1999). Whilethese tests are commonly termed “panel unit root” tests, theoretically, they are simplymultiple-series unit root tests that have been applied to panel data structures. Some ofthese approaches (like the Im–Pesaran–Shin or IPS test) view the panel data regressionas a system of N individual regressions and is based on the combination of independentDickey–Fuller tests for these N regressions. Generally, in this study, we have just two“truly” panel structured (firm specific) series. These are the dependent variable and theinvestment allocation index. Other series are macro variables that are assumed to becommon to all the firms, hence the regular unit root tests on the variables apply.

Unit root analysis and data poolability

The Eviews summary statistics for the various tests as applied to the dependent variable(It/Kt-1) and the investment allocation index are presented in Table 4. There is strong

statistical evidence to suggest that the dependent variable is stationary at level (thoughthe Breitung test tends to show that unit root exists at less than 10% significance level).However, unit root tests show that some of the independent variables are stationary atfirst difference while others are stationary at levels (see Table 4). Further, at between 5and 10% significance level, some of the I(1) variables can be regarded as I(0). Thissituation of having the dependent variable stationary and some explanatory variables,none stationary at first difference, could pose some estimation problems. We can getround this problem in some ways, which we discuss in the following section.

Generally, there are two traditional methods for estimating panel models: averagingand pooling. The former involves running N separate regressions and calculatingcoefficient means (see for example the mean group estimator method advanced by Pesaranand Smith, 1995). A drawback to averaging is that it does not account for the fact thatcertain parameters may be equal over cross sections. Alternatively, we could pool thedata and assume that the slope coefficients and error variances are identical.

25


Table 4: Panel unit root summary statisticsMethod Statistic Prob.** Cross sections Obs.

Null: Unit root (assumes common unit root process)

Levin, Lin & Chu t* -5.4232 0.0046 67 1594-1.2314 0.2142 67 1594

Breitung t-stat. 2.42531 0.0962 67 15980.56913 0.5311 67 1598

Null: Unit root (assumes individual unit root process)

Im, Pesaran and Shin W-Stat -47.156 0.0017 67 1592-3.6131 0.4233 67 1592

ADF-Fisher chi-square 24.7723 0.0052 67 16023.0124 0.2190 67 1602

PP-Fisher chi-square 36.9522 0.0082 67 16046.7132 0.3161 67 1604

Null: No unit root (assumes common unit root process)

Hadri Z-stat 1.3241 0.4263 67 16082.9512 0.0481 67 1608

Note: The foregoing is the panel unit root test summary for the endogenous variable. Sample is 1980 to 2003. Tests are in termsof individual effects (constant only). Automatic selection of lags is based on SIC (0 to 10) and Newey–West bandwidth selectionusing Bartlett Kernel. The first line of the statistics for each test is for the dependent variable, It /Kt-1, while the second line (below thefirst line) is for investment allocation index.** Probabilities for Fisher tests are computed using asymptotic chi-square distribution. All other tests assume asymptotic normality.

However, whereas there may be theoretical and empirical reasons to presume thatthe long-run coefficients are homogenous over the cross section, there are very fewpractical cases in which the short-run dynamics and error variances would behomogeneous, too. In the case of firm level investment, for example, it can be arguedthat differences in industrial and trade structure that sometimes exist between firms ofdifferent sectors would likely affect the short-run rather than long-run responses to tradepolicy changes (given their effects on the optimal life of capital goods).

Following this point, we consider it appropriate to use the Pesaran et al. (1999)pooled mean group estimation (PMGE) method, which is an intermediate case betweenthe averaging and pooling methods of estimation and involves aspects of both. Themethod restricts the long-run coefficients to be equal over the cross sections, but allowsfor the short-run coefficients and error variances to differ across cross sections. Thisimplies that we have to test for homogeneity

There are two popular approaches to examining poolability in panel data analysis.They are the joint Hausman test (see Pesaran et al., 1996, for details), and second alikelihood ratio approach. The Hausman test is based on the null of equivalence betweenthe PMG (pooled mean group) and MG (mean group) estimation. If we reject the nullwe reject homogeneity of our cross sections’ long-run coefficients, while significantstatistical difference between our two estimators would be indicative of panel mis-specification. The likelihood ratio test for long-run parameter heterogeneity is also widelyused and more conventional in this setting and has homogeneity as the null hypothesis(see Hsiao, 2003). In general, the tests require the estimation of the model under therestriction of common slopes across firms, as well as allowing heterogeneous slopes.This test is a generalization of the Chow test for the N linear regressions case, under thegeneral assumption of non spherical disturbances ε ∼ Ν (0, Ω ).


Table 5: Unit root test statistics for other explanatory variablesLevel 1st Diff

Variable ADF KPSS ADF KPSS Remark

∆ln (real GDP) -2.494 0.366 -5.847 0.337 I(0)/I(1)∆ln (user cost of capital) -1.915 0.303 -3.098 0.193 I(1)Bank credit as percentage of GDP 0.088 0.384 -5.430 0.398 I(1)Real GDP growth (unc.) -1.850 0.411 -3.041 0.270 I(1)Real exchange rate (unc.) -1.150 0.406 -3.680 0.288 I(1)M1 growth rate (unc.) -1.951 0.311 -3.062 0.201 I(1)Share fiscal deficit in GDP (unc.) -0.412 0.543 -2.911 0.321 I(1)Share current account in GDP (unc.) -2.412 0.332 -4.439 0.261 I(0)/I(1)Share external reserves in GDP (unc.) -1.409 0.490 -5.144 0.278 I(1)Average tariff (unc.) -0.886 0.493 -6.671 0.138 I(1)User cost of capital (unc.) -1.151 0.558 -3.314 0.181 I(1)Public expenditure on Infrastructure -0.413 0.481 -2.941 0.278 II(1)Restrictions on International capital -2.382 0.281 -3.651 0.179 I(0)/I(1)Dln real exchange rate -3.919 0.174 -5.618 0.284 I(0)Ratio (Export + Import) to GDP -1.144 0.527 -6.592 0.146 I(1)Import restriction: Tariff -2.413 0.315 -3.219 0.187 I(0)/I(1)Import restriction: cost of importing index -2.201 0.312 -3.961 0.211 I(0)/I(1)Black market premium -2.468 0.112 -5.157 0.078 I(0)/I(1)

Note: The optimal lag length for the ADF test is based on the Akaike information criterion with an upper lag of 4 and interceptonly. The approximate test critical values are -3.029 (5%) and -2.655 (10%). The KPSS test is based on Barlett KernelSpectral estimation method with automatic selection of the Newey–West bandwidth and intercept only. The approximate testcritical values are 0.4630 (5%) and 0.3470 (10%). The unc in parentheses in the first column implies the variables aremeasures of uncertainty.

A major reason for utilizing the PMGE in our empirical work is that the estimatedcoefficients in the model are not dependent upon whether the variables are I(1) or I(0).Pesaran and Shin (1998) present Monte Carlo evidence that the ARDL approach basedon the delta method can be reliably used in small samples to estimate and test hypotheseson the long-run coefficients in cases where there is a mixture of I(1) and I(0) regressors.

The test statistics (Table 6) show that our panel specification is statistically valid.This result is not too surprising since the explanatory macro variables are the same overthe years for all firms. The test of homogeneity across the various cross sections couldnot reject the null of homogeneity. Also, the chi-square test of random effect in the crosssection rejects the hypothesis that the long-run coefficients differ across the cross sections.

Table 6: Poolability tests

Test cross-section fixed effects

Effects test Statistic

Cross-section F(66, 1448) 0.942517 (0.7315)Cross-section χ2 (66) 3.988683 (0.6199)

Test cross-section random effects

Test summary χ2 (66) statistic 76.668125 (0.0047)


Estimation and interpretation of results

We carried out three panel regression estimations as reported in Table 7. EstimationsI and II are dynamic pooled mean group estimation without controlling for

unobserved firm-specific effects. Estimation II is derived after removing insignificantvariables (using 10% significance level) from estimation I. Estimation III is derivedfrom a fixed-effect estimation of II. Largely, discussion of the results will be based onthe estimates reported under III.

From Table 7, it can be observed that the one-period lag of investment ratio ispositively related to investment ratio. The relatively large autoregressive value of 0.6implies that investment ratio adjusts slowly to shocks in Nigeria. It can be observed thatamongst the conventional determinants of firm investment, both growth in real GDPand bank credit as a percentage of GDP positively affect investment ratio. The resultsindicate that in the short run if the economy grows at, say, 5% per annum, the rate ofcapital accumulation will be about 2.5% per annum. In the long run, such steady growthis capable of generating about 17% rate of capital accumulation.16 Furthermore, weobserved that current level of credit to the private sector (as a ratio of GDP) needs to bedoubled if a steady increase in investment rate of about 4% is to be achieved. In the setof volatility variables, we observed that volatility in real GDP, real exchange rate, tariffand user cost of capital were all negatively and significantly related to investment rate.

Table 7: Dynamic pool mean group regression results

Dependent variable ⇒ Net investment/capital (It/Kt-1)

Explanatory variables ⇓ (I) (II) (III)

Constant 0.0324 -0.0151(0.5618) (-0.3210)

One period lagged dependent 0.5152 0.6221 0.6177

variable (2.7634)** (3.0312)*** (2.9707)***

CF Dln (real GDP) 0.0634 0.0462 0.0522

conventional (3.0021)*** (2.8013)*** (2.9852)***factors Dln (user cost of capital) -0.0768

(0.9621)Bank credit as % of GDP 0.0323 0.0491 0.04121

(2.7640)*** (2.9341)*** (3.0650)***

UF,(measured in Real GDP growth -0.0712 -0.0234 -0.0202standard (-3.0231)*** (-2.7841)*** (-2.8754)**deviation)

-0.0076 -0.0097 -0.0109Real exchange rate (-2.6781)*** (-3.1211)*** (-3.1043)***

-0.0011M1 growth rate (-0.6431)

-0.0891Share fiscal deficit in GDP (-0.6710)

Share current account in GDP 0.0648(0.7851)

Share external reserves in GDP 0.0452 (0.7342)

Continued


Table 7, ContinuedAverage tariff -0.0621 -0.0033 -0.0064

(-2.8171)*** (-3.0121)*** (-3.1162)***

User cost of capital -0.0072 -0.0051 -0.0028(-2.7681)*** (-2.2521)** (-1.7651)

PF, Policy Public expenditure on infrastructure 0.0087factors (1.0016)

Restrictions on international capital 0.0363 0.0454 0.0411(2.6733)*** (2.9851)*** (3.0411)***

CR, Trade ∆ln real exchange rate -0.0064 -0.0073 -0.0092policy and (-3.2032)*** (-3.3411)*** (-2.9844)***credibility

Ratio (export+import) to GDP 0.0703 (1.7699)Import restriction: Tariff 0.0095 0.0094 0.0105

(2.9801)*** (2.1886)** (2. 5241)** Import restriction: Cost of importing 0.0124 0.0131 0.0205index (2.9078)*** (3.0652)*** (2.8429)***

-0.0604 -0.0312 -0.0279Black market premium (-2.7856)*** (-1.4521) (-1.0651)

0.0322 0.0131 0.0241No. bilateral investment treaties. (2.3227)** (1.4210) (1.6202)

Dummies & Year since membership of MIGA 0.0767interactions (0.6761)

Import intensity (with real exchange -0.0423 -0.0623 -0.0676rate) (-3.7588)*** (-3.3218)*** (-3.6411)*** Import intensity (with cost of -0.0526 -0.0602 -0.0771Importing) (-2.8957)*** (-3.0421)*** (-3.1318)***

Allocation index 0.0301(1.3425)

Allocation index (with dummy 0.0337 0.0426 0.0449>1985 =1) (2.9532)*** (2.8523)*** (2.8841)*** Exchange rate and sectoral 0.1121 0.1636 -0.2765dummy (2.8413)*** (2.9216)*** (-2.9763)***

No. of obs. 332 332 332Adjusted R2 0.77 0.73 0.71Durbin–Watson 2.09 2.11 1.97F-stat (p-value) 38.71 31.55 31.01

(0.000) (0.001) (0.001)

White heteroscedasticity-Consistent t-statistic in parentheses.***(**) Significance at 99(95)%.

Measured in standard deviation of the relevant variables, it can be observed that realGDP volatility has the highest coefficient (in absolute terms) followed by real exchangerate volatility. The interpretation is that investment reacts negatively to output volatilitymore than exchange rate volatility. Volatility in average tariff and user cost of capitalwere also found to be negatively related to investment rate. Looking at the size of thevolatility coefficients, we can infer that sustained growth in real output will tend toboost investment rate amongst firms than even a stable exchange rate regime. This isnot to suggest that a stable exchange rate is not important in boosting investment, ratherthe result shows a highly significant coefficient for output volatility. Again, of interest is


the high significance of the coefficient on tariff volatility. Though small in magnitude, itshows that tariff volatility of, say, one standard deviation can cause investment rate todecline by about 0.006 of a unit. In monetary terms, this could be enormous. In terms ofmacroeconomic volatility, we can infer that real output, exchange rate, average tariffand user cost of capital are important variables that policy makers should watch out for.

Amongst the trade policy and credibility variables, we observed that exchange ratedepreciation and black market premium had significant negative impact on investmentrate, while import restriction based on tariff index and import restriction based on costof import index had significant positive relationships with investment rate (this isexplained presently). The result indicates that on the average, in the short run, a 10%depreciation of the exchange rate can lead to about 0.1 unit contraction in investmentrate.

The coefficient on the tariff-based index of intermediate import restriction is slightlylarger than that of exchange rate (in absolute terms), implying that the effect of tariffadjustment on investment rate is larger than the impact of exchange rate depreciation oninvestment rate. Furthermore, the coefficient on the cost of importing index basedrestriction on import is larger that that of the tariff based and exchange rate depreciationin absolute terms. Note that the cost of import index is calibrated from 1 to 10 (see noteunder Table 1), with higher values implying reduced implicit cost of importing capitalgoods. A unit increase in the index (implying relaxation of import restriction) tends tohave more positive impact on capital accumulation than exchange rate depreciation.Another important observation is the relatively high and significant coefficient onrestrictions on international capital. The estimate suggests that a unit gain in the index(i.e., further relaxation of capital controls) can lead to about 0.04 unit increase ininvestment rate. Table 8 shows the result tends to suggest the importance of a liberalinternational capital regime in spurring investment in Nigeria. An estimate from theinteraction of the sectoral dummy with real exchange rate shows that the non-tradeablesector is more negatively affected by exchange rate adjustment than the tradeable sector.

Table 8 presents the results for the Tobin’s q model. The results are not too qualitativelydifferent in some respect from the results of the neoclassical model presented in Table7. It should be observed that the dependent variable in the Tobin’s q is the ratio of fixedinvestment to capital stock during the current period (and not past period). Second, thefirms included in the Tobin’s q model are only the quoted firms as we require stockmarket data in calculating the q ratio. Hence, the number of observations for the Tobin’sq is smaller. The results for the core Tobin’s q model in column 3 and for the extendedmodel in column 4.

The results from the estimate of the core Tobin’s q model indicate that the one-period lag of the dependent variable and the Tobin’s q are positively significant inexplaining firm investment amongst quoted firms in Nigeria. Based on the value of thelagged dependent variable (0.5), investments response to shock in the q model is similarto the neoclassical model presented in Table 7. The cash flow variable is also positivelysignificant at 90% significance level. To this core model, we append other variables. Itcan be observed that the one-period lag of the dependent variable, the Tobin’s q, and thecash flow variables are still positively significant in the presence of other variables.


Table 8: Regression estimates of the Tobin’s q model

Dependent variable ⇒(I/K)it

Explanatory variables ⇓ (I) (II) Constant 0.3771 (1. 7823) 0.0422 (0.5713) One-period lagged dependent variable 0.4653 (2.9871)*** 0.5166 (2.8660)***CF,conventional Tobin’s q 0.4029 (2.4351)** 0.5913 (2.5624)**factors One-period lag of cash flow to capital 0.2284 (1.7761)* 0.3616 (2.0788)**

UF (measured Real GDP growth -0.0141 (-2.8120)***in standard Real exchange rate -0.0078 (-3.0721)***deviation) M1 growth rate

Share fiscal deficit in GDPShare current account in GDPShare external reserves in GDPAverage tariff -0.0110 (-3.0864)***User cost of capital

PF, Policy Public expenditure on Infrastructurefactors Restrictions on International capital 0.0231 (2.9872)***

CR, Trade ∆ln real exchange rate -0.0431 (-3.0266)***policy and Ratio (export+import) to GDPcredibility Import restriction: Tariff 0.0231 (3.1127)***

Import restriction: Cost of importing index 0.0337 (3.6690)***Black market premiumNo. bilateral investment treaties.

Dummies Year since membership of MIGA& Import intensity (with real exchange rate) -0.0412 (-2.8776)***interactions Import intensity (with cost of importing) -0.0314 (-3.0721)***

Allocation index 0.2711 (1.9910)**Allocation index (with dummy >1985 =1) 0.0231 (3.0017)***Exchange rate and sectoral dummy -0.2301 (2.8128)***

No. of obs. 235 235Adjusted R2 0.32 0.66Durbin–Watson 2.27 1.93F-stat (p-value) 5.511 (0.0942) 24.61 (0.0021)

White heteroscedasticity-consistent t-statistic in parentheses.***(**)(*) Significance at 99(95)(90)%

Uncertainty in real output growth and the real exchange rate were observed to besignificant but with smaller coefficients (in absolute terms) than the estimates reportedin Table 7. The implication is that uncertainty in output growth and exchange rate affectsinvestment of unquoted firms more than quoted firms. A major reason could be thatmany of the quoted firms are subsidiaries of multinational corporations whose investmentdecisions are largely long term, unlike the unquoted firms, which are largely owned bylocal individuals. Again, we observed that uncertainty about user cost of capital was notsignificant in this model, while uncertainty about average tariff was observed to benegatively significant and with a larger coefficient (in absolute terms) than in theneoclassical model. In the Tobin’s q model, a black market exchange rate premium wasobserved not to be significant in investment decisions of quoted firms. This may bebecause the quoted firms have relatively easier access to foreign exchange from theofficial market. In terms of trade policy and credibility variables, it was observed thatreal exchange rate depreciation (-), tariff measure of import restrictions relaxation (+),


and cost of importing index (+), are the significant variables. The coefficients arerelatively larger (in absolute terms) for the quoted firms than for the total firm sample.The implication is that liberalizing the process of importing intermediate capital goodscan boost corporate investment in Nigeria.

In terms of the dummies and interactions, the results are not qualitatively differentfrom what obtains in the neoclassical model. The dummies on real exchange rate andcost of importing intermediate capital goods, when interacted with import intensity, arenegatively significant, suggesting that the exchange rate system negatively affects themore import intensive quoted firms more than the less import intensive quoted firms.When exchange rate is interacted with the (tradeable/non-tradeable) sectoral dummyclassification, we observed that the non-tradeable sector tends to be more negativelyand significantly affected by exchange rate policy than the tradeable sector.

These results are important in terms of trade policy credibility issue. Althoughexchange rate depreciation negatively affects investment, its depressing effect can becompensated for by a more liberal trade tariff regime and measures that reduce cost ofimportation (as it is observed that the magnitude of the tariff and import restrictioncoefficients is greater than that of exchange rate depreciation). This implies that whenthe government announces tariff reduction measures, it is important to stick to it over areasonable period of time as this will increase credibility and increase investment.Reversal of tariff reduction measures will harm investment as it is observed that tariffvolatility is negatively related to investment. Hence, a credible trade policy regime thatensures that an announced liberal policy will be sustained is likely to spur capitalaccumulation even in the presence of some exchange rate depreciation.

This is not to suggest that a highly variable exchange rate regime is investmentinducing, rather the result shows that real exchange rate volatility is negatively relatedto capital accumulation. The result suggests that a liberal trade tariff regime spursinvestment, however, and can even compensate for the adverse impact of exchange ratedepreciation on investment. But it is the case that over the past one decade or so, thegovernment of Nigeria has announced several liberal tariff regimes that were laterreversed. There are several cases in which goods being banned were later included inthe import list and then banned again. This kind of policy inconsistency and reversalwill definitely lead to lower policy credibility and hence reduction in investment. Fromthe results, this seems to have been the case in Nigeria.

In terms of dummies and other interactions, we observe that the import intensitydummy when interacted with the real exchange rate and cost of importing index showssignificant negative relationship with investment rate. The basic interpretation is thatfirms with high import intensity tend to react more to exchange rate depreciation andhigh cost of importing than firms with low import intensity. Unfortunately, most corporatefirms in Nigeria (as in our sample) are highly import intensive. This result may besuggestive of why we found that exchange rate depreciation and cost of importing capitalgoods have a negative impact on investment rate in Nigeria. Our measure of investmentallocation is also interacted with a dummy variable that measures the trade liberalizationregime in Nigeria. The dummy takes the value of unity for the period since 1985 andzero otherwise. The index is positively significant, which shows that trade liberalizationactually caused investment allocation to firms with higher returns. The result shows that


trade policy in Nigeria caused allocation of investment amongst firms in favour of firmswith higher efficiency of capital.

Implications of findings

Earlier in this study we described the practice of trade policy in Nigeria sinceindependence, as well as private investment response to the various trade episodes.

It was observed that trade policy gained prominence in Nigeria in the 1980s and thecommencement of extensive economic reform programme led to deregulation of variousprices and entry requirements into businesses. Various trade and exchange rate measureshave been introduced since the reform programme began with the aim of boostingproactive growth and investment activities in the non-oil sector. The response of corporateinvestment has been quite unimpressive. Why is this so?

The significance of private sector credit corroborates the argument in the literaturethat credit constraint is a major constraint facing domestic firms in Nigeria. Credit tomanufacturing firms has squeezed over the years because of high and variable inflation,low investment returns, and high returns from foreign exchange speculations. This iscorroborated by the negative coefficient on black market premiums in all the regressionestimates. Granting credit to manufacturing firms is seen as less profitable and risky inan environment characterized by uncertainty and large swings in relative prices.Speculative and other short-term investment activities become more profitable.

Furthermore, the results from our analyses show that a poor macroeconomicenvironment has significantly and negatively affected corporate investment in Nigeria.Uncertainty about the behaviour of key (relative) prices has had significant negativeimpact on investors’ perception of government's macroeconomic policy stance. Thevolatility of real exchange rate arising from the rapid and variable rates of depreciationhas caused significant uncertainty in the system. The study observed the negative andsignificant impact of volatility in interest rates, tariff rates and the cost of capital. Thebasic implication is that volatility in key prices has been a discouraging influence oncorporate investment. The macroeconomic environment has thus not been veryencouraging to investors.

In terms of the practice of trade policy, the results support the proposition that Nigeriahas significant potential for improving domestic capital accumulation if appropriatemeasures are put in place. We observed the significant but negative impact of tradepolicy (and related) variables such as exchange rate depreciation, import restriction andinternational capital flows on investment. Volatility in some of these variables – whichcan signal uncertainty and policy reversal – were observed to negatively affect domesticinvestment. This can be seen from the behaviour of the real exchange rate, the level oftariffs and their variability, which were observed by their signs to have significantlyslowed investment rates in Nigeria. Again, the cost of importing is seen to be prohibitivelyhigh.

The index used in this study (see notes under Table 1) shows that firms with highimport intensity tend to be more affected by the cost of importing. Although the resultsshow that the variable tends to be negatively significant for all firms, firms with highimport intensity are worse off. When combined with rising inflation and rapid devaluation


of the exchange rate, we can conclude that the cost of doing business is prohibitivelyhigh and discourages corporate investment. A policy variable that measures degree ofinternational capital mobility (see notes under Table 1) shows that the more liberal issuch a regime the higher will be the investment rate. Clearly, the study implies that tradepolicy has significant allocative effects in terms of investment amongst firms in Nigeria,with investment being allocated to firms with higher returns since liberalizationcommenced in 1986. What all this points at is that in the practice of trade policy, Nigeriawill need to further liberalize international transactions and ensure that the costs of suchtransactions are reduced drastically. This is important, since many of the firms havehigh import intensities.


6. Summary and conclusions

In an attempt to reduce dependence on oil exportation, the Government of Nigeriaover time has made efforts at increasing investment in other sectors of the economy,particularly manufacturing and agriculture. However, the poor response of investment

in these sectors has been widely recognized to be an important policy problem. In thisstudy firm-level evidence has been reported for investment in plant and equipment inthe manufacturing sector for 67 firms. We observed the negative impact ofmacroeconomic uncertainty, as measured by the volatility of key relative prices andother macro variables, on corporate investment in Nigeria. Furthermore, the studyobserved that trade policy practice in Nigeria has deterred investment by making thecost of importing high, which particularly affects firms with high import intensities.

Particularly, we observed that there are three main channels by which trade policypractice hinders corporate investment in Nigeria. First is through restrictions oninternational capital movement, the second is through the impact of rapid and variabledevaluation of the exchange rate, which has led to a more attractive speculative marketand lowers competitiveness of domestic firms and finally through import restrictions interms of tariff and cost of importing. It is observed that the practice of trade policy inNigeria, particularly liberalization, has led to investment allocation amongst firms. Thestudy calls for reduced restrictions on international capital controls and taking steps toreduce the cost of importing capital goods. Given the observed negative impact ofmacroeconomic uncertainty on corporate investment, the study calls for adoption of amore investment friendly macroeconomic environment by reducing uncertainty in thesystem.

35


Notes1. Weber (1991: 62) defines credibility as “the extent to which beliefs concerning a

policy conform to official announcements about this policy. ... Credibility maythus also be viewed as a measure of the degree to which policy makers tie theirhands on future policies by issuing policy announcements”. Hence, credibilitycan be regarded as the expectation that an announced policy will be carried out.

2. Trade liberalization in this study does not concentrate on tariffs alone, or even ontariffs and quantitative trade restrictions. Rather, the definition of liberalization,we believe, must encompass exchange rate policy and marketing arrangements aswell as the narrower range of policies that affect trade directly.

3. We consider trade reform as a set of economic instruments and institutionalarrangements that create an environment for traders (hence transactions) that ispredictable, consistent and sustainable in order to achieve better allocation ofresources.

4. See, for example, Busari and Fashanu (1998), Ikhide (1994), Atoyebi andOdedokun (1991), Emenuga (1996), Omoruyi (1995), and Soyibo (1996a/b).

5. The discussions and data used in this section are based on various issues of theAnnual Report and Statement of Accounts of the Central Bank of Nigeria. Furtherreading can be found in Olofin (1992) and Agbaje and Jerome (2004).

6. It should be observed that this idea started long before the Pindyck (1991)exposition. However, Pindyck's work could be described as a first attempt torigorously formalize the various ideas into a unified framework.

7. The authors wish to thank resource persons in group C at the AERC May 2005biannual for pointing this out.

8. The literature on real options theory is too vast to summarize here. Excellentsurveys are provided by Amran and Kulatilaka (1999), Copeland and Antikarov(2001), Coy (1999), Dixit and Pindyck (1994), and Lander and Pinches (1998).

9. This suggests that in modelling trade reversal, other factors like exchange rate,fiscal policy, etc., should be taken into consideration.

36


10. A classic exercise was also conducted for four African countries by Bigsten et al.(1997). Readers can consult this study to see several investment modelspecifications.

11. Interested readers are referred to Lee (2005) and Hamida (2000) for detaileddiscussions on the Tobin’s q.

12. The authors’ attention was drawn to this issue by resource persons at the May2005 AERC biannual workshop.

13. The authors wish to thank AERC resource persons for pointing this out.

14. Galindo et al. (2002) further discussed the advantages of using the sales-basedapproach.

15. The authors wish to acknowledge the contributions of resource persons in theAERC May 2006 biannual in bringing out these issues.

16. This is derived by dividing the coefficient on growth by one minus coefficient onthe one-period lag of investment rate.


ReferencesAbel, A.B. and J.C. Eberly. 1994. “A unified model of investment under uncertainty”. American

Economic Review, 84: 1369–84.Abel, A.B. and J.C. Eberly. 1997. “An exact solution for the investment and value of a firm facing

uncertainty, adjustment costs, and irreversibility”. Journal of Economic Dynamics and Control,21: 831–52.

Abel, A.B., A.K. Dixit, J.C. Eberly and R.S. Pindyck. 1996. “Options, the value of capital, andinvestment”. Quarterly Journal of Economics, 111: 753–77.

Agbaje, A. and A. Jerome. 2004.) “Institutional framework and the process of trade policy makingin African countries: The case of Nigeria”. Final report submitted to the African EconomicResearch Consortium, January.

Agenor, P.R. and P. Masson. 1999. “Credibility, reputation, and the Mexican peso crisis”. Journalof Money, Credit and Banking, 31(1): 70–84.

Aizenman, J. (1992) “Trade reforms, credibility, and development”. Journal of DevelopmentEconomics. 39(1): 163–87.

Aizenman, J. and N. Marion. 1993. “Policy uncertainty, persistence and growth”. Review ofInternational Economics, 2: 145–63.

Aizenman, J. and N. Marion. 1999. “Volatility and investment: Interpreting evidence fromdeveloping countries”. Economica, 66: 157–79.

Amran, M. and N. Kulatilaka. 1999. Real Options: Managing Strategic Investment in an UncertainWorld. Boston: Harvard Business School Press.

Arellano, R. 1990. “On the causes and effects of real exchange rate variability: A comparativeanalysis of the East Asian and Latin American experience”. Unpublished doctoral dissertation.University of California, Los Angeles.

Asiedu, E. 2002. “On the determinants of foreign direct investment to developing countries: IsAfrica different?” World Development, 30(1): 107–19.

Asiedu, E. 2004. “Policy reform and foreign direct investment in Africa: Absolute progress butrelative decline”. Development Policy Review, 22(1): 41–8.

Athukorala, P. and K. Sen. 1996. “Reforms and investment in India”. Departmental WorkingPapers No. 1996-06. Economics Research School of Pacific and Asian Studies (RSPAS),Australian National University.

Atoyebi, G.O. and M.O. Odedokun. 1991. “Determinants of sectoral fixed investment expenditureby foreign companies in Nigeria: An empirical approach”. Annals of the Social Science Councilof Nigeria, 3: 80–97.

Bardhan, P. 1984. The Political Economy of Development in India. Oxford: Basil Blackwell.Bassanini, Andrea. 2006. “Credibility, irreversibility of investment, and liberalization reforms in

LDCs”. Economic Notes, 35(2): 203–18.Ben-David, Dan. 1993. “Equalizing exchange: Trade liberalization and income convergence”.

Quarterly Journal of Economics, 108(3): 653–79.

38


Bertola, G. and R.J. Caballero. 1994. “Irreversibility and aggregate investment”. Review ofEconomic Studies. 61: 223–46.

Bhagwati, J. and T.N. Srinivasan. 1999. Outward-Orientation and Development: Are RevisionistsRight? Discussion Paper No. 806. Yale University Economic Growth Center, New Haven.

Bigsten, A., P. Collier, S. Dercon, B. Gauthier, J.W. Gunning, A. Isaksson, A. Oduro, R. Oostendorp,C. Pattillo, M. Soderbom, M. Sylvain, F. Teal and A. Zeufack. 1997. “Investment in Africa’smanufacturing sector: A four-country panel data analysis”. Working Paper Series, No. 59(WPS/97-11). The Centre for the Study of African Economies, University of Oxford.

Bleaney, Michael F. 1996. “Macroeconomic stability, investment, and growth in developingcountries”. Journal of Development Economics, 48: 461–77.

Bleaney, M. and D. Greenaway. 1993. “Adjustment to external imbalance and investment slumpsin developing countries”. European Economic Review, 37: 577–85.

Bleaney, M., D. Greenaway and D. Fielding. 1995. “Investment, trade liberalisation and structuraladjustment”. Journal of Development Studies, 32(2): 175–94.

Blejer, M. and M.S. Khan. 1984. “Government policy and private investment in developingcountries”. IMF Staff Papers, 31: 379–403.

Blomstrom, M., R. Lipsey and M. Zejan. 1996. “Is fixed investment the key to economic growth?”Quarterly Journal of Economics, 111: 269–76.

Breitung, J. 2000. “The local power of some unit root tests for panel data”. In B. Baltagi, ed.,Advances in Econometrics: Nonstationary Panels, Panel Cointegration and Dynamic Panels,pp. 161–77. Amsterdam: JAI Press.

Burnetti, A., G. Kisunko and B. Weder. 1997. “Credibility of rules and economic growth: Evidencefrom a worldwide survey of the private sector”. Policy Working Paper No. 1760. The WorldBank, Washington, D.C.

Busari, T.D. and F.O. Fashanu. 1998. “Macroeconomic policy regimes and private investment inNigeria: 1960–1994”. African Journal of Economic Policy, 5(1): 61–78.

Calvo, G. 1987. “On the cost of temporary policy”. Journal of Development Economics, 27:245–61.

Chen, Yu-Fu and M. Funke. 2003. “Option value, policy uncertainty, and the foreign directinvestment decision”. HWWA Discussion Paper No. 234. Hamburgisches Welt-Wirtschafts-Archiv (HWWA), Hamburg Institute of International Economics.

Chibber, A., M. Dailami and N. Shafik. 1992. Reviving Private Investment in DevelopingCountries: Empirical Studies and Policy Lessons. Amsterdam: North-Holland.

Chirinko, R. and H. Schaller. 1995. “Why does liquidity matter in investment equations?” Journalof Money, Credit, and Banking, 27: 527–48.

Choi, I. 2001. “Unit root tests for panel data”. Journal of International Money and Finance,20(2): 249–72.

Collier, P. 1995. “The marginalization of Africa”. International Labour Review, 134(4–5): 541–57.

Collier, P. and J.W. Gunning. 1994. “Trade and development: Protection, shocks, andliberalization”. In D. Greenway and L.A. Winters, eds., Surveys in International Trade. Oxford:Blackwell.

Collier, P. and J.W. Gunning. 1999. “Explaining African economic performance”. Journal ofEconomic Literature, 37(March): 64–111.

Collier, P. and C. Patillo. 2000. Investment and Risk in Africa. Basingstoke: Macmillan.Copeland, T. and V. Antikarov. 2001. Real Options – A Practitioner’s Guide. London: Texere

Publishing.Corbo, V. and S. Fischer. 1992. “Adjustment programs and bank support: Rationale and main

results”. In V. Corbo, S. Fischer and S.A.Webb., eds., Adjustment Lending Revisited: Policiesto Restore Growth. Washington, D.C.: The World Bank.

Corden, W. 1974. Trade Policy and Economic Welfare. Oxford: Clarendon Press.


Coy, R. 1999. “Exploiting uncertainty: The real-options revolution in decision making”. BusinessWeek, June 7: 118–24.

Cruces, G., P. Makdissi and Q. Wodon. 2003. “Poverty measurement under risk aversion usingpanel data”. http://www.usherbrooke.ca/economique/recherche/cahiers/2003/03_06_pm.pdf.

De Long, J. Bradford and Lawrence H. Summers. 1991. “Equipment investment and economicgrowth”. Quarterly Journal of Economics, 106(2): 445–502.

Dehn, Jan. 2000. “Private investment in developing countries: The effects of commodity shocksand uncertainty”. Working Paper Series No. 2000-11. Centre for the Study of AfricanEconomies, University of Oxford.

Dixit, A. and R. Pindyck. 1994. Investment under Uncertainty. Princeton: Princeton UniversityPress.

Dollar, D. 1992. “Outward-oriented developing economies really do grow more rapidly: Evidencefrom 95 LDCs, 1976–1985”. Economic Development and Cultural Change, 40(3): 523–44.

Drazen, A. and P. Masson. 1994. “Credibility of policies versus credibility of policymakers”.Quarterly Journal of Economics, 109: 735–54.

Eaton, J. and G. Grossman. 1985. “Tariffs as insurance: Optimal commercial policy when domesticmarkets are incomplete”. Canadian Journal of Economics, 18: 258–72.

Edwards, S. 1993. “Openness, trade liberalization, and growth in developing countries”. Journalof Economic Literature, 31(3): 1358–93.

Edwards, S. 1998. “Openness, productivity and growth: What do we really know?” EconomicJournal, 108: 383–98.

Emenuga, C. 1996. “Investment climate in Nigeria: An appraisal”. In Ademola Ariyo, ed., EconomicReform and Macroeconomic Management in Nigeria, pp. 39–54. Ibadan: Ibadan University Press.

Engel, C. and K. Kletzer. 1987. “Trade policy under endogenous credibility”. Center DiscussionPaper No. 543. Economic Growth Center, Yale University, September.

Everhart, S. and M. Sumlinski. 2001. Trends in Private Investment in Developing Countries.Statistics for 1970–2000 and the Impact on Private Investment of Corruption and the Qualityof Public Investment. IFC Discussion Paper No. 44. The World Bank, Washington, D.C.

Ferderer, J.P. 1993. “The Impact of uncertainty on aggregate investment – An empirical analysis”.Journal of Money, Credit and Banking, 25: 30–48.

Fielding, D. 1997. “Adjustment, trade policy and investment slumps: Evidence from Africa”.Journal of Development Economics, 52(1): 121–37.

Frankel, J. and D. Romer. 1999. “Does trade cause growth?” American Economic Review, 89(3):379–400.

Froot, Kenneth A. 1988. “Credibility, real interest rates, and the optimal speed of tradeliberalization”. Journal of International Economics, 25: 71–93.

Galindo, A., F. Schiantarelli and A. Weiss. 2002. “Does financial liberalization improve theallocation of investment?: Micro evidence from developing countries”. Working Paper No.467. Inter-American Development Bank.

Gertler, L.M. 1988. “Financial structure and aggregate economic activity: An overview”. Journalof Money, Credit and Banking, 20(3): 559–88.

Greenaway, D. and O. Morrissey. 1992. “Structural adjustment and liberalisation in developingcountries: What lessons have we learned?” Kyklos, 46: 241–61.

Greene, J. and D. Villanueva. 1991. “Private investment in developing countries: An empiricalanalysis”. IMF Staff Papers, 38(1): 33–58.

Grossman, G.M. and E. Helpman. 1991. Innovation and Growth in the Global Economy.Cambridge, Massachusetts: MIT Press.

Gunning, J.W. 1994. “Trade reform in Africa: The role of donors”. In J.W. Gunning, H. Kox, W. Timsand Y. de Wit., eds., Trade, Aid and Development, pp. 86–98. Basingstoke and London: Macmillan.


Hadjimichael, T.M. and D. Ghura. 1995. “Public policies and private savings and investment inSub-Saharan Africa: An empirical investigation”. Working Paper No. 95/19. InternationalMonetary Fund, Washington, D.C.

Hadjimichael, T.M., D. Ghura, M. Mühleisen, R. Nord and E. M. Uçer. 1995. Sub-Sahara Africa:Growth, Savings, and Investment, 1986–93. Occasional Paper No. 118. International MonetaryFund, Washington, D.C.

Hadri, K. 1999. “Testing the null hypothesis of stationarity against the alternative of a unit root inpanel data with serially correlated errors”. The Econometrics Journal, 3: 148–61.

Hamida, L.B. 2000. “The study of Swiss investment expenditures using Tobin’s q model 1948–1995”. www.ecomod.org/files/papers/1604.pdf. Accessed on 20 July 2006.

Harrigan, J. and P. Mosley. 1991. “Evaluating the impact of World Bank structural adjustmentlending”. Journal of Development Studies, 27: 95–134.

Harrison, Ann. 1996. “Openness and growth. A time-series, cross-country analysis for developingcountries”. Journal of Development Economics, 48: 419–47.

Hassett, K.A. and G.E. Metcalf. 1999. “Investment with uncertain tax policy: Does random taxpolicy discourage investment?” Economic Journal, 109: 372–93.

Hayashi, F. 1982. “Tobin’s average q and marginal q: A neoclassical interpretation”. Econometrica,50: 213–24.

Hsiao, C. 2003. Analysis of Panel Data, 2nd edition. Cambridge: Cambridge University Press.Ibarra, L.A. 1995. “Credibility of trade policy reform and investment: The Mexican experience”.

Journal of Development Economics, 47: 39–60.Ikhide, S.I. 1994. “External shocks, savings, investment and credit availability: The evidence

from Nigeria”. African Journal of Economic Policy, 1(2): 35–45.Im, K.S., M.H. Pesaran and Y. Shin. 2003. “Testing for unit roots in heterogeneous panels”.

Journal of Econometrics, 115: 53–74.Ingersoll, J. and S.A. Ross. 1992. “Waiting to invest: Investment and uncertainty”. Journal of

Business, 65: 1–29.Jorgenson, D. 1963. “Capital theory and investment behavior”. American Economic Review, 53:

247–59.Jorgenson, D.W. 1971. “Econometric studies of investment behavior: A survey”. Journal of

Economic Literature, IX(4): 1111–47.Lander, D.M. and G.E. Pinches. 1998. “Challenges to the practical implementation of modeling

and valuing real options”. Quarterly Review of Economics and Finance, 38: 537–67.Lee, H. 2005. “The impact of uncertainty on investment: Empirical evidence from manufacturing

firms in Korea”. http://faculty.washington.edu/karyiu/confer/sea05/papers/lee_hy.pdf.Accessed 20 July 2006.

Lemi, A. and S. Asefa. 2001. “Foreign direct investment and uncertainty: The case of Africaneconomies”. Proceedings of the International Business and Economics Research Conference,Reno, Nevada, 8–12 October 2001.

Lin and C. Chu. 2002. “Unit root tests in panel data: Asymptotic and finite-sample properties”.Journal of Econometrics, 108: 1—24.

Levine, R. and D. Renelt. 1992. “A sensitivity analysis of cross-country growth regression”.American Economic Review, 82: 942–63.

Loungani, P. and M. Rush. 1995. “The effects of changes in reserve requirements on investmentand GNP”. Journal of Money, Credit and Banking, 27(2): 511–26.

Maddala, G. and S. Wu. 1999. “A comparative study of unit root tests with panel data and a newsimple test”. Oxford Bulletin of Economics and Statistics, 61(S1): 631–52.

McCulloch, R. and M.F. McPherson. 2002. “Promoting and Sustaining Trade and Exchange Reformin Africa: An Analytical Framework”. In Malcolm F. McPherson, ed., Promoting and SustainingTrade and Exchange Rate Reform in Africa, pp. 15–54. Cambridge, Massachusetts: EAGER(Equity and Growth through Economic Research).


McDonald, R. and D. Siegel. 1985. “Investment and the valuation of firms when there is anoption to shut down”. International Economic Review, 26: 331–49.

McDonald, R. and D. Siegel. 1986. “The value of waiting to invest”. Quarterly Journal ofEconomics, 101: 707–27.

McKinnon, R.I. 1973. Money and Capital in Economic Development. Washington, D.C.: TheBrookings Institute.

Metcalf, Gilbert E. and Donald Rosenthal. 1995. “The ‘new’ view of investment decisions andpublic policy analysis: An application to green lights and cold refrigerators”. Journal of PolicyAnalysis and Management, 14(4): 517–31.

Michaely, M., D. Papageorgiou and A.M. Choski. 1991. Liberalizing Foreign Trade, Vol. 7:Lessons of Experience from the Developing World. Cambridge, Massachusetts: BasilBlackwell.

Mosley, P., J. Harrigan and J. Toye. 1991. Aid and Power: The World Bank and Policy-basedLending. London: Routledge.

Ng, F. and A. Yeats. 1997. “Open economies work better! Did Africa’s protectionist policiescause its marginalization in world trade?” World Development, 25(6): 889–904.

OECD. 2001. The Development Dimensions of Trade. Policy Brief (October), Organization forEconomic Cooperation and Development, Paris.

Olofin, S. 1992. “Trade policies in Nigeria”. In D. Salvatore, ed., National Trade Policies: Studiesin Comparative Economic Policies, Vol. 2. Amsterdam: North Holland.

Omoruyi S.E. 1995. “Trade, investment, and debt accumulation”. Debt Trend, 1(September):103–18.

Pesaran, M.H. and Y. Shin. 1998. “An autoregressive distributed-lag modelling approach tocointegration analysis”. In S. Strøm, ed., Econometrics and Economic Theory in the 20thCentury: The Ragnar Frisch Centennial Symposium. Cambridge: Cambridge University Press.

Pesaran, M.H., Y. Shin and R. Smith. 1999. “Pooled mean group estimation of dynamicheterogeneous panels”. Journal of the American Statistical Association, 94: 621–34.

Pesaran, M.H. and R. Smith. 1995. “Estimating long-run relationships from dynamic heterogeneouspanels”. Journal of Econometrics, 68: 79–113.

Pesaran, M.H., R. Smith and K.S. Im. 1996. “Dynamic linear models for heterogeneous panels”.In L. Matyas and P. Sevestre, eds., The Econometrics of Panel Data. Dordrecht: KluwerAcademic Publishers.

Pessoa, S.A., S.M. Pessoa and R. Rob. 2004. “Elasticity of substitution between capital andlabour: A panel data approach”. www.ie.ufrj.br/.../pesquisa/elasticity_of_substitution_between_capital_and_labor_a_panel_data_approach.pdf

Pindyck, R. 1991. “Irreversibility, uncertainty, and investment”. Journal of Economic Literature.29: 1110–49.

Pindyck, R.S. and A. Solimano. 1993. “Economic instability and aggregate investment”. NBERMacroeconomics Annual, 8: 259–319.

Price, Simon. 1995. “Aggregate uncertainty, capacity utilization and manufacturing investment”.Applied Economics, 27: 147–54.

Rajan, R. and S. Marwah. 1998. “The effects of policy uncertainty on the choice and timing offoreign direct investment: An exploratory firm-level analysis”. Journal of EconomicDevelopment, 23(1): 37–56.

Rodrik, D. 1989a. “Credibility of trade policy reform: A policy maker’s guide”. World Economy,12(1): 1–16.

Rodrik, D. 1989b. “Promises, promises: Credible policy reform via signaling”. Economic Journal,99: 756–72.

Rodrik, D. 1990. “How should structural adjustment programs be designed?” World Development,18(7): 933–47.


Rodrik, D. 1991. “Policy uncertainty and private investment in developing countries”. Journal ofDevelopment Economics, 36(2): 229–42.

Rodrik, D. 1992. “The limits of trade policy reform in developing countries”. Journal of EconomicPerspectives, 6(1): 87–105.

Rodrik, D. 1997. “Why is trade reform so difficult in Africa?” Paper Prepared for the PlenarySession of the AERC Bi-nnual Research Workshop, Nairobi, May.

Romer, P. 1994. “New goods, old theory, and the welfare costs of trade restrictions”. Journal ofDevelopment Economics, 43(1): 5–38.

Ruggiero, J. 2003. “Data envelopment analysis with stochastic data”. http://www.sba.udayton.edu/research/working_papers/wp10.pdf. Accessed 5 May 2005.

Sachs, J. and A. Warner. 1995. “Economic reforms and the process of global integration”. BrookingsPapers on Economic Activity, 2: 1–118.

Serven, L. 1996. “Irreversibility, uncertainty and private investment: Analytical issues and somelessons for Africa”. Journal of African Economies, 6(3): 229–68.

Serven, L. 2002. “Real exchange rate uncertainty and private investment in developing countries”.Policy Research Working Paper No. 2823. Washington, D.C.: The World Bank.

Serven, L. and A. Solimano. 1993. “Economic adjustment and investment performance indeveloping countries: The experience in the 1980s”. In L. Serven and A. Solimano, eds.,Striving for Growth after Adjustment: The Role of Capital Formation. The World Bank,Washington, D.C.

Shaw, E. 1973. Financial Deepening in Economic Development. Oxford: Oxford UniversityPress.

Soyibo, A. 1996a. The Savings–Investment Process in Nigeria: An Empirical Study of the SupplySide. Research Paper No. 12. African Economic Research Consortium, Nairobi.

Soyibo, A. 1996b. The Transmission of Savings to Investment in Nigeria. Research Paper No. 48.African Economic Research Consortium, Nairobi.

Staiger, R. and G. Tabellini. 1987. “Discretionary trade policy and excessive protection”. AmericanEconomic Review. 77: 823–37.

Tobin, J. 1969. “A general equilibrium approach to monetary theory”. Journal of Money, Creditand Banking, 1: 15–29.

Van Wijnbergen, S. 1985. Aid, Export Promotion and the Real Exchange Rate: An AfricanDilemma. Discussion Paper No. 88. Centre for Economic Policy Research, London.

Weber, A.A. 1991. “Reputation and credibility in the European monetary system”. EconomicPolicy, 12: 57–102.

Winters, L.A. 2001. “Trade policies for poverty alleviation in developing countries”. In B.Hoekman, P. English and A. Mattoo, eds., Trade Policy, Economic Development andMultilateral Negotiations: A Sourcebook. Washington, D.C.: The World Bank.

World Bank. 1988. Adjustment Lending. PPR Policy and Research Papers No. 1. The WorldBank, Washington, D.C.


Appendix – Data descriptionTable A1: Industrial distribution of sampled firms

Firm category Number of firms

1 Chemicals and paints 52 Conglomerates 63 Food, beverages & tobacco 74 Industrial/domestic products 95 Breweries 46 Construction 67 Packaging 68 Healthcare 89 Automobiles and tyres 210 Cement & building materials 611 Footwear 112 Textile 313 Machinery marketing 214 Printing and publishing 2

Total sampled firms 67

For the selected firms, we obtained a balanced sample over the specified period.Equipment, plants and machinery were regarded as fixed corporate capital and net changesin the values of these assets are regarded as investment. Stock market capitalization dataare as reported by the Securities and Exchange Commission, while for simplicity weuse market value for the replacement cost of capital when calculating the Tobin’s q. Inother words, we do not account for technical progress at any particular year of calculation.

For Table A2, growth in GDP is measured as ∆lnyt where ∆ represents discrete timechange, and exchange rate depreciation is measured as ∆lnet. The real exchange rate ismeasured as nominal domestic price of a unit of USA currency multiplier by ratio ofdomestic price level to USA price level. The consumer price index (CPI) is used toproxy for price level. Hence, an increase in the value of the real exchange rate impliesincreased international competitiveness for Nigeria. The black market premium ismeasured as the simple difference between the price of currency at the parallel marketand the price at the official market, using the US dollar as the currency of reference.

44


Table A2: Descriptive statistics of variables Mean Median Max. Min. Std. Skew. Kurtosis Jarque- Prob.

dev. Bera

Growth in realGDP a 1.77 3.87 5.63 -7.39 5.23 -1.35 3.07 1.53 0.47Real GDP growthvolatility 4.31 2.53 14.03 0.24 5.67 1.21 2.85 1.23 0.54Real exch. ratedepreciationa 0.38 0.47 0.63 0.16 0.20 -0.08 1.43 0.52 0.77Real exc. ratedepreciationvolatility 0.32 0.29 0.76 0.05 0.27 0.94 2.65 0.76 0.68Nom. exch. ratedepreciationa 0.22 0.22 0.45 0.05 0.16 0.35 1.81 0.40 0.82Nom. exch. ratedepreciationvolatility 0.24 0.21 0.57 0.02 0.22 0.62 2.11 0.48 0.79M1 volatility 10.83 14.22 15.38 3.15 5.60 -0.53 1.47 0.72 0.70Fiscal deficitvolatility 2.59 2.73 4.62 1.13 1.34 0.52 2.15 0.37 0.83Current accountbalance volatility 8.75 6.22 15.37 5.85 4.10 0.95 2.29 0.85 0.65Nominal interestrate volatility 4.33 3.04 6.98 2.05 2.30 0.33 1.25 0.73 0.69Inflation volatility 14.70 18.32 20.61 4.35 7.07 -0.63 1.74 0.67 0.72External reservevolatility 3.91 4.02 5.28 1.68 1.50 -0.52 1.94 0.46 0.80Average tariff 15.52 12.70 24.98 7.85 7.15 0.34 1.52 0.55 0.76Tariff volatility 2.85 2.16 5.41 1.13 1.71 0.62 1.97 0.54 0.76Cost of capital -2.80 -4.70 6.43 -8.17 5.56 0.99 2.66 0.85 0.65Cost of capitalvolatility 12.33 14.46 17.64 5.09 5.31 -0.43 1.55 0.59 0.74Ratio of (export +import) to GDP a 52.70 57.11 70.44 33.83 16.42 -0.19 1.31 0.62 0.73Black marketpremiuma 137.98 130.10 239.49 39.34 82.30 0.07 1.54 0.45 0.80Cost of importindexb 4.64 4.60 5.90 3.40 1.03 0.03 1.53 0.45 0.80Infrastructureexpenditurea 3.55 3.07 5.86 1.61 2.06 0.20 1.25 0.67 0.71Share of privatesector credit inGDP a 9.65 6.84 21.66 5.14 6.79 1.42 3.15 1.68 0.43Internationalcapital restrictionindexb 3.02 3.10 5.40 0.00 2.04 -0.42 2.14 0.30 0.86Investmenttreatiesc 5 3 15 0 6.285 0.844 2.260 0.71 0.70MIGAc 0.8 1 1 0 0.447 -1.500 3.250 1.89 0.39

Notes:a. Data sourced from Statistical Bulletin of the Central Bank of Nigeria, various issues.b. Data published by the Fraser Institute (see footnotes 13 and 14).c. Compiled from www.unctad.org/en/docs/poiteiiad2.en.pdf, www.worldbank.org/icsid/treaties/main.htm, andhotdocs.usitc.gov/pub3741/Table1-9.pdf (all accessed by 27 May 2006). Volatility measures are calculatedby authors.


Other recent publications in the AERC Research Papers Series:

External Aid Inflows and the Real Exchange Rate in Ghana, by Harry A. Sackey, Research Paper 110.Formal and Informal Intitutions’ Lending Policies and Access to Credit by Small-Scale Enterprises in Kenya:

An Empirical Assessment, by Rosemary Atieno, Research Paper 111.Financial Sector Reform, Macroeconomic Instability and the Order of Economic Liberalization: The Evidence

from Nigeria, by Sylvanus I. Ikhinda and Abayomi A. Alawode, Research Paper 112.The Second Economy and Tax Yield in Malawi, by C. Chipeta, Research Paper 113.Promoting Export Diversification in Cameroon: Toward Which Products? by Lydie T. Bamou, Research Paper 114.Asset Pricing and Information Efficiency of the Ghana Stock Market, by Kofi A. Osei, Research Paper 115.An Examination of the Sources of Economic Growth in Cameroon, by Aloysius Ajab Amin, Research Paper 116.Trade Liberalization and Technology Acquisition in the Manufacturing Sector: Evidence from Nigeria, by

Ayonrinde Folasade, Research Paper 117.Total Factor Productivity in Kenya: The Links with Trade Policy, by Joseph Onjala, Research Paper 118.Kenya Airways: A Case Study of Privatization, by Samuel Oyieke, Research Paper 119.Determinants of Agricultural Exports: The Case of Cameroon, by Daniel Gbetnkon and Sunday A. Khan,

Research Paper 120.Macroeconomic Modelling and Economic Policy Making: A Survey of Experiences in Africa, by Charles

Soludo, Research Paper 121.Determinants of Regional Poverty in Uganda, by Francis Okurut, Jonathan Odwee and Asaf Adebua, Research

Paper 122.Exchange Rate Policy and the Parallel Market for Foreign Currency in Burundi, by Janvier D. Nkurunziza,

Research Paper 123.Structural Adjustment, Poverty and Economic Growth: An Analysis for Kenya, by Jane Kabubo-Mariara and

Tabitha W. Kiriti, Research Paper 124.Liberalization and Implicit Government Finances in Sierra Leone, by Victor A.B. Davis, Research Paper 125.Productivity, Market Structure and Trade Liberalization in Nigeria, by Adeola F. Adenikinju and Louis N.

Chete, Research Paper 126.Productivity Growth in Nigerian Manufacturing and Its Correlation to Trade Policy Regimes/Indexes (1962–

1985), by Louis N. Chete and Adeola F. Adenikinju, Research Paper 127.Financial Liberalization and Its Implications for the Domestic Financial System: The Case of Uganda, by

Louis A. Kasekende and Michael Atingi-Ego, Research Paper 128.Public Enterprise Reform in Nigeria: Evidence from the Telecommunications Industry, by Afeikhena Jerome,

Research Paper 129.Food Security and Child Nutrition Status among Urban Poor Households in Uganda: Implications for Poverty

Alleviation, by Sarah Nakabo-Sswanyana, Research Paper 130.Tax Reforms and Revenue Mobilization in Kenya, by Moses Kinyanjui Muriithi and Eliud Dismas Moyi,

Research Paper 131.Wage Determination and the Gender Wage Gap in Kenya: Any Evidence of Gender Discrimination? by Jane

Kabubo-Mariara, Research Paper 132.Trade Reform and Efficiency in Cameroon’s Manufacturing Industries, by Ousmanou Njikam, Research Paper 133.Efficiency of Microenterprises in the Nigerian Economy, by Igbekele A. Ajibefun and Adebiyi G. Daramola,

Research Paper 134.The Impact of Foreign Aid on Public Expenditure: The Case of Kenya, by James Njeru, Research Paper 135.The Effects of Trade Liberalization on Productive Efficiency: Electrical Industry in Cameroon, by

Ousmanou Njikam, Research Paper 136.How Tied Aid Affects the Cost of Aid-Funded Projects in Ghana, by Barfour Osei, Research Paper 137.Exchange Rate Regimes and Inflation in Tanzania, by Longinus Rutasitara, Research Paper 138.Private Returns to Higher Education in Nigeria, by O.B. Okuwa, Research Paper 139.Uganda’ s Equilibrium Real Exchange Rate and Its Implications for Non-Traditional Export Performance,

by Michael Atingi-Ego and Rachel Kaggwa Sebudde, Research Paper 140.Dynamic Inter-Links among the Exchange Rate, Price Level and Terms of Trade in a Managed Floating

Exchange Rate System: The Case of Ghana, by Vijay K. Bhasin, Research Paper 141.Financial Deepening, Economic Growth and Development: Evidence from Selected Sub-Saharan African

Countries, by John E. Udo Ndebbio, Research Paper 142.The Determinants of Inflation in South Africa: An Econometric Analysis, by Oludele A. Akinboade, Franz K.

Siebrits and Elizabeth W. Niedermeier, Research Paper 143.


The Cost of Aid Tying to Ghana, by Barfour Osei, Research Paper 144.A Positive and Normative Analysis of Bank Supervision in Nigeria, by A. Soyibo, S.O. Alashi and M.K.

Ahmad, Research Paper 145.The Determinants of the Real Exchange Rate in Zambia, by Kombe O. Mungule, Research Paper 146.An Evaluation of the Viability of a Single Monetary Zone in ECOWAS, by Olawale Ogunkola, Research Paper 147.Analysis of the Cost of Infrastructure Failures in a Developing Economy: The Case of the Electricity Sector in

Nigeria, by Adeola Adenikinju, Research Paper 148.Corporate Governance Mechanisms and Firm Financial Performance in Nigeria, by Ahmadu Sanda, Aminu

S. Mikailu and Tukur Garba, Research Paper 149.Female Labour Force Participation in Ghana: The Effects of Education, by Harry A. Sackey, Research Paper 150.The Integration of Nigeria's Rural and Urban Foodstuffs Market, by Rosemary Okoh and P.C. Egbon,

Research Paper 151.Determinants of Technical Efficiency Differentials amongst Small- and Medium-Scale Farmers in Uganda: A

Case of Tobacco Growers, by Marios Obwona, Research Paper 152.Land Conservation in Kenya: The Role of Property Rights, by Jane Kabubo-Mariara, Research Paper 153.Technical Efficiency Differentials in Rice Production Technologies in Nigeria, by Olorunfemi Ogundele and

Victor Okoruwa, Research Paper 154.The Determinants of Health Care Demand in Uganda: The Case Study of Lira District, Northern Uganda, by

Jonathan Odwee, Francis Okurut and Asaf Adebua, Research Paper 155.Incidence and Determinants of Child Labour in Nigeria: Implications for Poverty Alleviation, by Benjamin C.

Okpukpara and Ngozi Odurukwe, Research Paper 156.Female Participation in the Labour Market: The Case of the Informal Sector in Kenya, by Rosemary Atieno,

Research Paper 157.The Impact of Migrant Remittances on Household Welfare in Ghana, by Peter Quartey, Research Paper 158.Food Production in Zambia: The Impact of Selected Structural Adjustment Policies, by Muacinga C.H. Simatele,

Research Paper 159.Poverty, Inequality and Welfare Effects of Trade Liberalization in Côte d’ Ivoire: A Computable General Equilibrium

Model Analysis, by Bédia F. Aka, Research Paper 160.The Distribution of Expenditure Tax Burden before and after Tax Reform: The Case of Cameroon, by Tabi

Atemnkeng Johennes, Atabongawung Joseph Nju and Afeani Azia Theresia, Research Paper 161.Macroeconomic and Distributional Consequences of Energy Supply Shocks in Nigeria, by Adeola F. Adenikinju

and Niyi Falobi, Research Paper 162.Analysis of Factors Affecting the Technical Efficiency of Arabica Coffee Producers in Cameroon, by

Amadou Nchare, Research Paper 163.Fiscal Policy and Poverty Alleviation: Some Policy Options for Nigeria, by Benneth O. Obi, Research Paper 164.FDI and Economic Growth: Evidence from Nigeria, by Adeolu B. Ayanwale, Research Paper 165.An Econometric Analysis of Capital Flight from Nigeria: A Portfolio Approach, by Akanni Lawanson, Research

Paper 166.Extent and Determinants of Child Labour in Uganda, by Tom Mwebaze, Research Paper 167.Implications of Rainfall Shocks for Household Income and Consumption in Uganda, by John B. Asiimwe

and Paul Mpuga, Research Paper 168.A Modelling of Ghana’s Inflation Experience: 1960–2003, by Mathew K. Ocran, Research Paper 169.Oil Wealth and Economic Growth in Oil Exporting African Countries, by Olomola P. Akanni, Research

Paper 170.Relative Price Variability and Inflation: Evidence from the Agricultural Sector in Nigeria, by Obasi O.

Ukoha, Research Paper 171.Sources of Technical Efficiency among Smallholder Maize Farmers in Southern Malawi, by Ephraim

Chirwa, Research Paper 172.The Determinants of School Attendance and Attainment in Ghana: A Gender Perspective, by Harry A.

Sackey, Research Paper 173.Private Returns to Education in Ghana: Implications for Investments in Schooling and Migration, by

Harry A. Sackey, Research Paper 174.Privatization and Enterprise Performance in Nigeria: Case Study of Some Privatized Enterprises, by

Afeikhena Jerome, Research Paper 175.Determinants of the Capital Structure of Ghanian Firms, by Joshua Abor, Research Paper 176.


THIS RESEARCH STUDY was supported by a grant from the African EconomicResearch Consortium. The findings, opinions and recommendations are those of theauthors, however, and do not necessarily reflect the views of the Consortium, itsindividual members or the AERC Secretariat.

Published by: The African Economic Research ConsortiumP. O. Box 62882 - City SquareNairobi 00200, Kenya

Printed by: Modern Lithographic (K) LtdP. O. Box 52810 - City SquareNairobi 00200, Kenya

ISBN 9966-778-25-X

© 2008, African Economic Research Consortium.

ContentsList of tablesList of figuresAbstractAcknowledgements

1. Introduction 1

2. Characterizing Nigeria’s trade policy regimes 5

3. Review of the literature 10

4. Methodological framework 16

5. Analyses and results 25

6. Summary and conclusions 35

Notes 36

References 38

Appendix 44

List of tables1. Description and measurement of the explanatory variables 182. Correlation of measures of volatility 203. Expected impact of exchange rate depreciation on investment 234. Panel unit root summary statistics 265. Unit root test statistics for other explanatory variables 276. Poolability tests 287. Dynamic pool mean group regression results 298. Regression estimates of the Tobin’s q model 31

A1. Industrial distribution of sampled firms 44A2. Descriptive statistics of variables 45

List of figures1. Average tariff rate 62. Ratio of aggregate investment to total and non oil GDP 63. (Export + import) as a ratio of GDP 64. A schema on credibility concept 15

AbstractThis study presents an empirical assessment of the impact of trade policy practice andits credibility on private investment using firm level data of 67 Nigerian firms over theperiod 1980–2003. The study draws a distinction between sample variability anduncertainty and constructs measures of the volatility of innovations to key trade andmacroeconomic variables. It then examines their association with private corporateinvestment by adding these constructed measures to an empirical investment equationthat is estimated using panel data econometric method. The results underscore therobustness of the links among private investment, trade policy and macroeconomicuncertainty. Many of the trade and volatility measures considered in the paper showstrong negative association with private investment. Furthermore, the study observedthat trade policy practice in Nigeria has deterred investment by making the cost ofimporting high, which particularly affects firms with high import intensity. Three mainchannels are identified through which trade policy practice hinders corporate investmentin Nigeria. First is through restrictions on international capital movement. The secondis through the impact of rapid and variable devaluation of the exchange rate, which hasled to a more attractive speculative market and lowed the competitiveness of domesticfirms, and finally through import restrictions in terms of tariff and cost of importing. Inaddition, the negative impact of real exchange rate uncertainty on investment issignificantly larger in firms that are import intensive. The study argues, among otherthings, that there is the need to liberalize international capital movement in Nigeria anddrastically reduce the cost of importing capital goods.

Keywords: Private investment, trade policy credibility, macroeconomic uncertainty, paneldata estimation, Nigeria

JEL classification: C23, E22, F13, R42

AcknowledgementsWe are immensely thankful to the African Economic Research Consortium (AERC) forfunding this study. We also express our gratitude and appreciation to the resource personsand researchers in group C of the AERC Biannual Research Workshops during the periodin which we undertook this study. All enduring errors and omissions are solely andentirely our responsibility.

Documents

Private Investment Behaviour and Trade Policy Practice in ... · PRIVATE INVESTMENT BEHAVIOUR AND TRADE POLICY PRACTICE IN NIGERIA 3 of the most important factors determining the